claude-video-enhance-audio — AI Audio Enhancement

This sub-skill handles AI-model-based audio processing. For FFmpeg-native audio operations (loudnorm, EQ, compression, silence removal, mixing), use claude-video-audio instead.

Pre-Flight

1. Activate venv: source ~/.video-skill/bin/activate
2. Check free VRAM: nvidia-smi --query-gpu=memory.free --format=csv,noheader,nounits
3. Verify required tools are installed for the requested operation

Source Separation (Demucs v4)

Isolate vocals, drums, bass, or other instruments from audio/video.