Update README.md
Browse files
README.md
CHANGED
|
@@ -77,7 +77,7 @@ Launch the included Gradio interface:
|
|
| 77 |
python danish_tts.py
|
| 78 |
```
|
| 79 |
|
| 80 |
-
Access at `http://localhost:7860` for an interactive TTS experience.
|
| 81 |
|
| 82 |
## Voice Control
|
| 83 |
|
|
@@ -166,20 +166,12 @@ The included `danish_tts.py` provides a comprehensive web interface with:
|
|
| 166 |
- Requires base model `sesame/csm-1b` to function
|
| 167 |
- Voice control limited to male/female binary selection
|
| 168 |
|
| 169 |
-
## Technical Details
|
| 170 |
-
|
| 171 |
### Model Architecture
|
| 172 |
- **Base**: CSM-1B encoder-decoder with depth decoder
|
| 173 |
- **Audio Format**: 24kHz, generated via audio tokens
|
| 174 |
- **LoRA Integration**: Language projections only, speaker/codec frozen
|
| 175 |
- **Memory Requirements**: ~8GB VRAM for inference
|
| 176 |
|
| 177 |
-
### Files Included
|
| 178 |
-
- LoRA adapter weights
|
| 179 |
-
- Processor configuration
|
| 180 |
-
- Gradio web interface (`danish_tts.py`)
|
| 181 |
-
- Training scripts and utilities
|
| 182 |
-
|
| 183 |
## Citation
|
| 184 |
|
| 185 |
If you use this model, please cite:
|
|
|
|
| 77 |
python danish_tts.py
|
| 78 |
```
|
| 79 |
|
| 80 |
+
Access at `http://localhost:7860` for an interactive TTS experience. Or use the live [Huggingface Space](https://huggingface.co/spaces/nicolajreck/csm-1b-danish-tts-space).
|
| 81 |
|
| 82 |
## Voice Control
|
| 83 |
|
|
|
|
| 166 |
- Requires base model `sesame/csm-1b` to function
|
| 167 |
- Voice control limited to male/female binary selection
|
| 168 |
|
|
|
|
|
|
|
| 169 |
### Model Architecture
|
| 170 |
- **Base**: CSM-1B encoder-decoder with depth decoder
|
| 171 |
- **Audio Format**: 24kHz, generated via audio tokens
|
| 172 |
- **LoRA Integration**: Language projections only, speaker/codec frozen
|
| 173 |
- **Memory Requirements**: ~8GB VRAM for inference
|
| 174 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 175 |
## Citation
|
| 176 |
|
| 177 |
If you use this model, please cite:
|