01 / Drop
Choose a song
Any song or a music video, up to 15 minutes.
Free tool
Take the singing out of a song for karaoke, a cover or practice, and get the vocals alone as well. An AI model trained to separate music runs in this tab, so your song is never uploaded. Both tracks download as MP3 or WAV. No sign-up, no watermark.
This runs in your browser. Your files are never uploaded.
How it works
01 / Drop
Any song or a music video, up to 15 minutes.
02 / Separate
The voice and the music are pulled apart, about 8 seconds at a time.
03 / Save
The instrumental for karaoke and the vocals on their own.
Questions
No. The AI model is downloaded to your browser, from Hugging Face, the open model library, and runs there, so the file never leaves your computer. No account and no limit on the number of songs.
HTDemucs, the music separation model by Meta AI researchers, released under the MIT licence, in a browser version by the demucs-web project. It splits a song into vocals, drums, bass and the rest.
Good, but not perfect: a trace of the voice can stay in the instrumental, especially reverb and backing vocals, and bright instruments can leak into the vocals. In our test, a voice over made-up music, the separated voice was about 11 dB cleaner than in the mix. We have not measured it on real songs, so listen before you use it.
It depends on your computer. With WebGPU, which uses the graphics card, our test song of 20 seconds took about 5 seconds; without it, the processor is much slower. The first time, the model downloads, 181 MB. Songs up to 15 minutes.
For practice, karaoke at home or your own songs, yes. To publish a remix or a cover of someone else's song you need the rights holder's permission.
Yes, with remove music from video, which keeps the picture and the voices.