By ShinobiTools Team · Last updated: July 2026
Pull a ready-to-drop acapella out of any track: the isolated vocal as a clean MP3 for mashups, edits and remixes. Free, no account, no watermark.
Split your track free →Producers used to hunt forums for a leaked acapella or settle for a phasey "vocal only" that still had the beat ghosting under it. StemGrab lifts the vocal straight out of the commercial mix, so you get a usable acapella from the exact version you want, not whatever happened to leak.
Vocals, drums and bass come out remarkably clean on most modern, well-mixed tracks. Guitar and piano are newer, experimental stems and can be rougher, especially in dense mixes; lo-fi, live or heavily reverbed recordings are harder for any tool. Give the tricky ones a listen before you build on them.
For the plain "just remove the singing" angle see vocal remover, and if you want the words rather than the audio, extract vocals from a song covers the same stem. Clearing a mashup for release? Read stems & the law first.
An acapella on its own is a curiosity. An acapella that sits on top of someone else's instrumental is a mashup, and whether that works comes down to two numbers: key and tempo. Get them wrong and no amount of EQ will save it, because a vocal a semitone away from the backing track is the one mistake a listener hears instantly, even if they can't name it.
StemGrab measures both and shows them beside the finished stems, so run both songs through the tool, not just the one you're taking the vocal from. Write down the two keys and the two BPMs and you can see the job before you start it. Same key, tempos within a few percent: drop it in and nudge. A tempo gap under about 6% will stretch cleanly with any modern time-stretch algorithm. Beyond that you start hearing artefacts on consonants, and beyond about 10% the vocal begins to sound seasick.
Keys that are a fifth apart, or a relative major and minor (A minor and C major, for instance), often work with no transposition at all because they share most of their notes. A semitone or two apart is a straight transpose. Anything further and you're better off finding a different instrumental than fighting it, because transposing a vocal more than about three semitones starts to change the character of the voice.
Separation quality varies by source material, and knowing what hurts saves you from re-uploading the same song three times. Feed the tool the best copy you have: a lossless file or a high-bitrate MP3 rather than something ripped from a video at 96 kbps, because the artefacts in a low-bitrate file get amplified by separation rather than hidden by it. Avoid anything with an intro DJ drop or a radio-edit voiceover on top; that voice is a vocal too, and it will come out fused with the one you want.
Three things show up in almost every extracted vocal, and all three are normal. Bleed is a faint trace of the instrumental behind the voice, most audible in quiet passages, and it disappears the moment you put the acapella over a new backing track. Reverb tails can sound clipped, because the model lifts the voice out while the room it was sung in stays partly behind; if the vocal sounds oddly dry, adding a touch of your own reverb makes it sit properly. And ad-libs panned hard left or right sometimes end up quieter than the lead, which you can lift with automation if you need them.
Before you audition anything, use the mixer on the results page: solo the vocal stem, jump to the second chorus (the busiest part of most songs) and listen there. If it holds up in the chorus, the verses will be fine.
On most modern mixes, yes — the model separates the vocal rather than cancelling frequencies, so there's little to no beat under it. Dense mixes can leave faint bleed.
The acapella keeps the song's original key and tempo. Time-stretch or pitch-shift it in your DAW if your project needs a different one.
192 kbps MP3, with the instrumental and other stems available from the same result, plus a zip of everything.
The tool is free to use, but the rights to the recording stay with whoever owns them. Practising privately is always fine; for anything you publish — a remix, cover or mashup — get permission first. See stems & the law.