stemrollerapp/stemroller
Desktop app that uses Facebook's Demucs neural network to separate songs into vocal and instrumental stems.
Not currently ranked — collecting fresh signals.
star history
StemRoller is a cross-platform desktop application that leverages Facebook’s Demucs deep learning algorithm to perform audio source separation. Users search for songs via YouTube integration and the app extracts individual stems such as vocals, drums, bass, and other instruments. The frontend is built with Svelte while the backend runs Python inference using the Demucs model.
Frequently asked
- What is stemrollerapp/stemroller?
- Desktop app that uses Facebook's Demucs neural network to separate songs into vocal and instrumental stems.
- Is stemroller open source?
- Yes — stemrollerapp/stemroller is an open-source project tracked on heatdrop.
- What language is stemroller written in?
- stemrollerapp/stemroller is primarily written in Svelte.
- How popular is stemroller?
- stemrollerapp/stemroller has 3.1k stars on GitHub.
- Where can I find stemroller?
- stemrollerapp/stemroller is on GitHub at https://github.com/stemrollerapp/stemroller.