Any-to-Any
Transformers
Safetensors
English
Korean
raon
image-feature-extraction
speech
audio
multimodal
custom_code
Instructions to use KRAFTON/Raon-Speech-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use KRAFTON/Raon-Speech-9B with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("KRAFTON/Raon-Speech-9B", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update README
Browse files
README.md
CHANGED
|
@@ -48,6 +48,10 @@ It trains on millions of hours of English-Korean speech-text datasets with the f
|
|
| 48 |
|
| 49 |
## Benchmark Results
|
| 50 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
Measured with LibriSpeech test-clean samples on single-GPU setups via streaming TTS. All values are averaged.
|
| 52 |
|
| 53 |
| Metric | RTX 6000 Pro | L40S |
|
|
|
|
| 48 |
|
| 49 |
## Benchmark Results
|
| 50 |
|
| 51 |
+
<div align="center">
|
| 52 |
+
<img src="assets/raon-speech-speechchat.png" alt="Raon-Speech Benchmark Results" width="800">
|
| 53 |
+
</div>
|
| 54 |
+
|
| 55 |
Measured with LibriSpeech test-clean samples on single-GPU setups via streaming TTS. All values are averaged.
|
| 56 |
|
| 57 |
| Metric | RTX 6000 Pro | L40S |
|