v2.0.0 — Docker, Web UI & Multi-Source Support

v2.0.0 — Major Release

Features

• AI-powered stem separation using Meta's Demucs models (4-stem and 6-stem)
• YouTube support — download and process videos/playlists directly
• Spotify support — process Spotify tracks via spotdl
• Docker containerization — CPU and GPU profiles, zero-dependency setup
• Web UI — browser-based interface for easy use
• Batch processing — process entire directories
• Custom remixing — mix stems with custom volume levels
• Audio preview — preview stems before saving
• GPU acceleration — automatic CUDA detection
• Multiple output formats — WAV, MP3, FLAC, OGG, AAC
• Metadata preservation — retains ID3 tags from source files
• Config file — save preferences in ~/.stem-separator.yaml

Install

pip install -e .
# or
docker compose --profile cpu up -d