| transformers # for model and tokenizer | |
| torch # PyTorch for running the model | |
| numpy # for audio data handling | |
| livekit == 1.0.6 # Use a compatible version of livekit | |
| torchaudio # for audio processing with PyTorch | |
| pydub # for audio file manipulation (optional, depending on use case) | |
| livekit-agents>=0.12.17 |