Load a song and the AI splits it into vocals, drums, bass, guitar and piano. Mute the vocals, export, and you have an instrumental. Everything runs on your PC — your audio is never uploaded.
No song-length caps, no daily limits, no watermark, no account.
Your audio stays on your machine. Local files separate with no internet at all.
4-stem (vocals, drums, bass, other) or 6-stem (adds guitar and piano) models.
Key and speed change, loop a section and a metronome on the same screen.
It uses current open AI separation models. Some vocal reverb can remain on certain songs, but it is usually more than enough for karaoke or practice tracks.
It depends on your PC. A graphics card (GPU) makes it much faster, and it also runs on CPU.
Yes. Mute everything except vocals, or choose per-stem export to get the vocal file on its own.
Not yet — Windows 10 and 11 only. There is also an Android app.
Use it for personal practice and study. Check the original copyright before publishing or selling anything you make.