Free AI-powered music demixer
Effortlessly split your songs into individual stems - vocals, bass, drums, guitar, piano, and more - using the best and latest AI models in our free web application. It runs locally on your computer 🏽, ensuring unlimited usage without any restrictions. Powered by the Demucs Hybrid Transformer AI models, created and maintained by Sevag H.
Runs locally and privately in your browser!
Unlike similar products, it’s free to use and doesn’t store your data. All processing is done in your browser, and your files are never uploaded anywhere to fully respect user privacy. It runs well on computers and very slowly on smartphones; user beware.
Support this site!
There is a MAX MEMORY option to speed up demixing and support longer songs!🚀🔥 We have tested up to 32 GB in Firefox and 16 GB in Chrome. Read the beginner’s tutorial if you need help.
- Demucs 4-source stems: vocals, drums, bass, melody
- Demucs 6-source stems: vocals, drums, bass, piano, guitar, other
DEMIX SONGS INTO STEMS
Disclaimers and updates
- ⚠️ We take code quality seriously! For bugs or requests, open a GitHub issue, or message us on Instagram or by e-mail
- 💻 If you experience crashes, try a different MAX MEMORY setting (lower or higher!)
- ⚖️ 📄 Model weights and outputs are governed by the Meta Research license
- Input files can be almost any audio format, but the outputs are always stereo wav files @ 44100 Hz
Check the GitHub releases page for information on recent updates and improvements to the site!