Pull the acapella out of any song
Built for the remix. Upload a record and ChordSonic's pipeline hands back the vocal on its own — same tempo, same pitch, same length as the master — ready to drop onto a timeline under a new beat.
Free · no signup · daily limit applies
Feed it the record
MP3 or WAV, up to 15 MB (25 MB with a free account) — the finished master is exactly what this wants, no need to hunt for a clean version. Uploads made without an account stay tied to your browser session and are cleared within a day.
The vocal comes off the master
ChordSonic's pipeline separates the performance from the production around it, at the original sample positions — so the acapella lines up with the record bar for bar.
Into the DAW
Preview it in the browser, download it as an MP3, and take the instrumental of the same separation too if the plan is a mashup rather than a remix.
One separation, both halves of a mashup
- Vocals arrive ticked on their own — press download and that file is the acapella.
- Now untick vocals and tick everything else, download again: that is the instrumental of the same record.
- Anything in between is legal too — vocal plus drums for a stripped edit, vocal plus bass for a dubplate feel.
An acapella is only half the information.
The file tells you nothing about what it was sung over. Lay it on a bed a semitone away and it will sound sour before the first bar is out; lay it on a tempo two BPM off and it drifts by the second verse. Both facts are already in the record, and the same upload can be made to give them up.
Run the track through ChordSonic's analysis and you get the key, the tempo and the whole chord progression on a timeline — so you know what to write under the voice, not just what pitch to avoid. DJs who mix by code can convert that key on the Camelot wheel and read off everything that mixes with it.
What the acapella will sound like. Honest version: it is cut out of a master, so reverb tails, printed delays and a little cymbal bleed ride along with it, and the loudest, densest records leave the most behind. That is normally fine under a new arrangement and obvious when soloed dry.
Came for something slightly different?
These four pages run the same separation — what changes is which parts arrive ticked and what the page is written to help you do next. Nothing is lost by landing on the wrong one, but the right one saves you the reading.
The same separation without the remix framing. If you only want the vocal file and none of the mixing advice, start there instead — it is one click shorter.
For a spoken voice rather than a sung one — a voiceover over a music bed, an interview, a sample with someone talking on it.
The inverse: keeps the music, drops the voice. That is the other half of a mashup, and the same upload gives you both.
When you want every part out, not just the voice — drums to sample, bass to re-play, the lot.
Acapella extractor FAQ
How do I get the acapella from a song?
Upload the track and leave the selection as it arrives — vocals only. ChordSonic's pipeline lifts the vocal performance out of the finished master and the download is that performance alone, at the length and timing of the original, so it drops straight onto a timeline against the record it came from.
Will the acapella be in time and in key for a remix?
It is cut from the master, so it keeps the original tempo and pitch exactly — nothing is stretched or resampled. What you still need is the numbers: the same upload can be run through ChordSonic's analysis to get the key, the tempo and the full chord progression, which is what tells you whether your new instrumental will sit under it or fight it.
Can I get the instrumental of the same track for a mashup?
Yes, and from the same separation — no second upload and no second charge. In the result panel untick vocals and tick everything else, then download again: that mix is the instrumental. A mashup usually wants both halves of one song, and this is how you take them.
Why can I still hear reverb or a faint beat under the vocal?
Because the reverb tail and the delay throws were printed onto the vocal when the record was mixed — they are part of the performance now, not a layer sitting over it. Dense, loud masters leave the most bleed. For a remix this is usually survivable: a high-pass and your own reverb over the top tend to hide what is left.
Do the ad-libs, doubles and backing vocals come out separately?
No. Separation works by instrument, not by take, so the lead, the doubled chorus and every ad-lib arrive together on one vocal track. If you want the lead alone you are back to editing by hand — but you are editing an acapella instead of a full mix, which is the hard part solved.
Am I allowed to release a remix made from this?
That is a rights question, not a technical one, and it does not change because the acapella was made by software. Bootlegs, edits and remixes of a record you do not own still need the owner's permission to be released or monetised. Extracting an acapella for practice, for a DJ set you are not selling, or for a remix you intend to clear is the usual honest ground.
Do I need an account, and does the paid plan give a better acapella?
No account for one track a day — no signup, no card. And no: the vocal gets a track of its own in the standard 4-part split that everyone gets, so a paid plan does not buy you a cleaner acapella. What the Supporter and Pro plans add is the 6-part split, which puts guitar and piano on their own tracks — useful for rebuilding an instrumental, irrelevant to the vocal itself.
Create your own music
Generate unique new tracks of your own — built from the key, tempo and chords of any song you upload.
Find the key and tempo the vocal was sung to
The same file, read instead of cut: key, BPM and every chord change on a play-along timeline — the numbers your new instrumental has to agree with.
Analyze a trackMixing by code? Convert it on the Camelot wheel.