About NoVocal

NoVocal is a browser-based tool for musicians, singers, DJs and video creators. Upload a song and an AI separation model splits it into six stems — vocals, drums, bass, guitar, piano and other instruments — so you can make a karaoke track, pull out an acapella, or practise along with the band.

How it works

Separation runs on hosted GPUs using Demucs, an open-source music source separation model originally published by Meta AI Research. Your browser uploads the file directly to storage, the model separates it, and the stems stream back to the player on the page. Mixing and WAV export happen in your browser.

Your files

Uploads and stems are deleted automatically after 24 hours. We do not listen to, share or train on your audio. Please only upload audio you have the right to use; separating a song does not change who owns it.

Contact

Questions, bugs or copyright notices: email us.