Update README.md
Browse filesClean up whitespaces
README.md
CHANGED
|
@@ -238,9 +238,8 @@ pipeline_tag: automatic-speech-recognition
|
|
| 238 |
|
| 239 |
# 🦜Parakeet-unified-en-0.6b: Unified ASR model for offline and streaming inference
|
| 240 |
|
| 241 |
-
| [Model architecture](#model-architecture)
|
| 242 |
-
|
|
| 243 |
-
| [Language](#datasets)
|
| 244 |
|
| 245 |
Parakeet-unified-en-0.6b is an English automatic speech recognition (ASR) model based on transducer architecture (RNN-T) combining both offline and streaming inference (up to 160ms latency) in one model. It is trained on the ASRSet dataset, which contains approximately 250,000 hours of US English (en-US) speech across diverse acoustic conditions. The model transcribes speech to English alphabet, spaces, and apostrophes with punctuation and captalization support.
|
| 246 |
|
|
@@ -368,22 +367,16 @@ We would recommend to use the following context parameters for different latenci
|
|
| 368 |
### Input
|
| 369 |
|
| 370 |
- Input Type(s): Audio
|
| 371 |
-
|
| 372 |
- Input Format(s): wav
|
| 373 |
-
|
| 374 |
- Input Parameters: One-Dimensional (1D)
|
| 375 |
-
|
| 376 |
- Other Properties Related to Input: Maximum Length in seconds specific to GPU Memory, No Pre-Processing Needed, Mono channel is required. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.
|
| 377 |
|
| 378 |
|
| 379 |
### Output
|
| 380 |
|
| 381 |
- Output Type(s): Text String in English
|
| 382 |
-
|
| 383 |
- Output Format(s): String
|
| 384 |
-
|
| 385 |
- Output Parameters: One-Dimensional (1D)
|
| 386 |
-
|
| 387 |
- Other Properties Related to Output: No Maximum Character Length, transcribe punctuation and capitalization. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.
|
| 388 |
|
| 389 |
|
|
|
|
| 238 |
|
| 239 |
# 🦜Parakeet-unified-en-0.6b: Unified ASR model for offline and streaming inference
|
| 240 |
|
| 241 |
+
| [Model architecture](#model-architecture) | [Model size](#model-architecture) | [Language](#datasets) |
|
| 242 |
+
|---|---|---|
|
|
|
|
| 243 |
|
| 244 |
Parakeet-unified-en-0.6b is an English automatic speech recognition (ASR) model based on transducer architecture (RNN-T) combining both offline and streaming inference (up to 160ms latency) in one model. It is trained on the ASRSet dataset, which contains approximately 250,000 hours of US English (en-US) speech across diverse acoustic conditions. The model transcribes speech to English alphabet, spaces, and apostrophes with punctuation and captalization support.
|
| 245 |
|
|
|
|
| 367 |
### Input
|
| 368 |
|
| 369 |
- Input Type(s): Audio
|
|
|
|
| 370 |
- Input Format(s): wav
|
|
|
|
| 371 |
- Input Parameters: One-Dimensional (1D)
|
|
|
|
| 372 |
- Other Properties Related to Input: Maximum Length in seconds specific to GPU Memory, No Pre-Processing Needed, Mono channel is required. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.
|
| 373 |
|
| 374 |
|
| 375 |
### Output
|
| 376 |
|
| 377 |
- Output Type(s): Text String in English
|
|
|
|
| 378 |
- Output Format(s): String
|
|
|
|
| 379 |
- Output Parameters: One-Dimensional (1D)
|
|
|
|
| 380 |
- Other Properties Related to Output: No Maximum Character Length, transcribe punctuation and capitalization. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.
|
| 381 |
|
| 382 |
|