Music Analyzer
Comprehensive music file analysis: lyrics extraction via vocal separation + Whisper, audio features via librosa, and Claude interprets the raw data into genre, mood, instruments etc.
Prerequisites
- Python 3.10+ with PyTorch, torchaudio, openai-whisper, librosa, soundfile
- ffmpeg must be in PATH
- Scripts located in:
scripts/(relative to this skill's base directory) - NVIDIA GPU with CUDA support recommended (works on CPU too, just slower)
- A
requirements.txtis inc
[Description truncada. Veja o README completo no GitHub.]