Instructions to use kyutai/moshiko-pytorch-bf16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Moshi
How to use kyutai/moshiko-pytorch-bf16 with Moshi:
# pip install moshi # Run the interactive web server python -m moshi.server --hf-repo "kyutai/moshiko-pytorch-bf16" # Then open https://localhost:8998 in your browser
# pip install moshi import torch from moshi.models import loaders # Load checkpoint info from HuggingFace checkpoint = loaders.CheckpointInfo.from_hf_repo("kyutai/moshiko-pytorch-bf16") # Load the Mimi audio codec mimi = checkpoint.get_mimi(device="cuda") mimi.set_num_codebooks(8) # Encode audio (24kHz, mono) wav = torch.randn(1, 1, 24000 * 10) # [batch, channels, samples] with torch.no_grad(): codes = mimi.encode(wav.cuda()) decoded = mimi.decode(codes) - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -91,8 +91,8 @@ The training was performed on 127 DGX nodes provided by Scaleway, accounting for
|
|
| 91 |
## Citation
|
| 92 |
|
| 93 |
```
|
| 94 |
-
@techreport{
|
| 95 |
-
author = {Alexandre D\'efossez
|
| 96 |
title = {Moshi: a speech-text foundation model for real-time dialogue},
|
| 97 |
institution = {Kyutai},
|
| 98 |
year={2024},
|
|
|
|
| 91 |
## Citation
|
| 92 |
|
| 93 |
```
|
| 94 |
+
@techreport{kyutai2024moshi,
|
| 95 |
+
author = {Alexandre D\'efossez and Laurent Mazar\'e and Manu Orsini and Am\'elie Royer and Patrick P\'erez and Herv\'e J\'egou and Edouard Grave and Neil Zeghidour},
|
| 96 |
title = {Moshi: a speech-text foundation model for real-time dialogue},
|
| 97 |
institution = {Kyutai},
|
| 98 |
year={2024},
|