igna/LuxTTS
1
0
Fork 0
Code Activity

Update README with model loading instructions and MPS support

Browse Source
This commit is contained in:
Yatharth Sharma
2026-01-28 16:40:07 -05:00
committed by GitHub
parent 2f2668a55e
commit 34820963ee
1 changed files with 10 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
README.md
View File

@@ -40,7 +40,15 @@ pip install -r requirements.txt
#### Load model: #### Load model:
```python ```python
from zipvoice.luxvoice import LuxTTS from zipvoice.luxvoice import LuxTTS
lux_tts = LuxTTS('YatharthS/LuxTTS', device='cuda', threads=2) ## change device to cpu for cpu usage
# load model on GPU
lux_tts = LuxTTS('YatharthS/LuxTTS', device='cuda')
# load model on CPU
# lux_tts = LuxTTS('YatharthS/LuxTTS', device='cpu', threads=2)
# load model on MPS for macs
# lux_tts = LuxTTS('YatharthS/LuxTTS', device='mps')
``` ```
#### Simple inference #### Simple inference
@@ -119,7 +127,7 @@ A: Yes, currently it uses float32. Float16 should be significantly faster(almost
- [x] Release model and code - [x] Release model and code
- [x] Huggingface spaces demo - [x] Huggingface spaces demo
- [ ] Release mps support - [x] Release MPS support (thanks to @builtbybasit)
- [ ] Release code for float16 inference - [ ] Release code for float16 inference
## Acknowledgments ## Acknowledgments