Split any song into the voice and the music. You get both files — the instrumental without the singing, and the acapella without the band — and if you want to go further, a second pass splits the lead from the backing vocals. The first minutes are free.
Stem separator
Split a song into vocals and instrumental
Start from a song, come away with the voice on its own and the instrumental on its own.
Previous runs
What you get back
Two files from one song: the instrumental, which is the whole arrangement with the singing taken out, and the vocals, which is everything sung with the music taken away. Tick Also split lead and backing vocals and the vocal file is split again into the main line and the harmony stack — a second pass most vocal removers do not offer, and the part of this tool people come back for.
What people do with the files
Karaoke and live covers — the instrumental is the backing track, from the actual record rather than a MIDI imitation.
Practice and transcription — hear the vocal line on its own, or play along without the original singer in the way.
Sampling and remixing — either stem is raw material.
AI covers — convert the isolated vocal to another voice and put it back over the untouched instrumental. Separating first is what keeps the music free of artifacts.
What decides the quality
The mix does. A voice sitting clearly on top of the music separates cleanly; a voice buried under heavy layers, distortion or wall-to-wall pads shares its space with the band, and the seams show. Dense choruses are the hardest case, verses the easiest. If the vocal comes out with the room still on it, the reverb remover takes that off as a second step.
How it works
A source-separation model has learned, from an enormous amount of music, what voices sound like and what instruments sound like, and reconstructs each from the mixture. It is the same family of models behind every serious stem splitter; the lead-and-backing second pass is the part that is rarer, and it is explained on its own page.
Common questions
Is it free?
Ten free minutes on the stem separator, and separate free minutes on the noise remover and reverb remover. They do not share a pool. One job costs the length of its audio once, and every output of that job downloads at no extra cost. A subscription removes the limits.
What files can I use?
WAV, MP3, FLAC, M4A and OGG, one or many at a time. Output in the format you choose, including lossless WAV or FLAC.
Do I get the instrumental and the vocals both?
Yes, always. One job produces every stem, and each downloads separately.
Whose music can I use?
Work with audio you have the rights to use. What you make from someone else's record is between you and the rights holder; the tool does not change that. See the acceptable use policy.
Do my files stay private?
Files are processed for your job and are not shared. See the privacy policy for retention details.