| # Base dependencies - install first for better caching | |
| numpy>=1.24.0 | |
| pandas>=2.0.0 | |
| scikit-learn>=1.3.0 | |
| # Audio processing | |
| librosa>=0.10.0 | |
| soundfile>=0.12.0 | |
| # ML frameworks | |
| # PyTorch 2.2+ with CUDA support (required for Flash Attention on A10G GPUs) | |
| torch>=2.2.0 | |
| torchaudio>=2.2.0 | |
| # torchcodec required for audio decoding in datasets library | |
| torchcodec>=0.1.0,<1.0.0 | |
| transformers>=4.30.0 | |
| # Hugging Face ecosystem | |
| huggingface_hub>=0.20.0 | |
| datasets>=2.14.0 | |
| accelerate>=0.26.0 # Required for Trainer with PyTorch | |
| # UI | |
| gradio>=5.49.1 | |