StemDeck: Free Open-Source Local Stem Splitter by Thales Pereira
StemDeck is a free, open-source stem separation application developed by solo developer Thales Pereira for Windows, macOS, and Linux. It runs entirely on your machine — no subscription, no cloud upload, no account required — and was built so Pereira's kid could practice drums by isolating parts from existing tracks. The tool is available directly from the developer's GitHub repository.
Features and formats
- Local inference using pre-trained models (no internet needed after download)
- Splits audio into stems: vocals, drums, bass, and other
- Supports common audio formats: WAV, FLAC, MP3, OGG
- Drag-and-drop interface with progress indication
- Batch processing for multiple files
- MIT license — royalty-free for any use, including commercial
System requirements
- Windows 10/11 (x64)
- macOS 11+ (Apple Silicon and Intel)
- Linux (AppImage, Debian/Ubuntu packages)
- ~2 GB RAM minimum; 4 GB+ recommended for larger files
- No GPU required, but CUDA/Metal acceleration supported where available
Quick facts
StemDeck is completely free and open-source under the MIT license with no time limits, sign-ups, or feature gates. Download the latest release for your platform from the GitHub repository and run the installer or AppImage — no DAW integration needed, it works as a standalone app.
FAQ
Is StemDeck really free with no hidden costs? Yes. The application and its source code are released under the MIT license, meaning you can use, modify, and distribute it freely, including in commercial projects.
What audio formats does StemDeck support for input and output? It accepts WAV, FLAC, MP3, and OGG files and writes separated stems as WAV files.
Does StemDeck require an internet connection to work? No. All processing happens locally on your computer after the initial download; no cloud services or accounts are involved.