tipi/kyutai
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Laurent Mazare 2025-06-16 21:39:36 +02:00 committed by GitHub
parent 08ba5dae03
commit 2f5e913852
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 23 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
README.md
View File

@ -1,2 +1,25 @@
# delayed-streams-modeling # delayed-streams-modeling
Delayed Streams Modeling (DSM) is a flexible formulation for streaming, multimodal sequence-to-sequence learning. Delayed Streams Modeling (DSM) is a flexible formulation for streaming, multimodal sequence-to-sequence learning.
## Speech To Text
### PyTorch implementation
```bash
python -m moshi.run_inference --hf-repo kyutai/stt input.mp3
```
### MLX implementation
```bash
python -m moshi_mlx.run_inference --hf-repo kyutai/stt-mlx ~/tmp/bria-24khz.mp3 --temp 0
```
## License
The present code is provided under the MIT license for the Python parts, and Apache license for the Rust backend.
The web client code is provided under the MIT license.
Note that parts of this code is based on [AudioCraft](https://github.com/facebookresearch/audiocraft), released under
the MIT license.
The weights for the models are released under the CC-BY 4.0 license.