Instructions to use kyutai/stt-2.6b-en with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Moshi
How to use kyutai/stt-2.6b-en with Moshi:
# pip install moshi # Run the interactive web server python -m moshi.server --hf-repo "kyutai/stt-2.6b-en" # 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/stt-2.6b-en") # 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 library name
#3
by nielsr HF Staff - opened
Given the model is compatible with Transformers, this PR updates the library_name, ensuring a "Use this model" button appears at the top right with a sample code snippet.
Hi, thanks for the PR. We had some issues with supporting the both the Transformers version and our PyTorch version in the same Hugging Face model repo so we're in the process of migrating the Hugging Face version to a separate one - probably called kyutai/stt-2.6b-en-hf. We'll need to make the change there.
vvolhejn changed pull request status to closed