Stem Separation
Pull any song apart, and keep every detail
Split a finished track into vocals, drums, bass and the rest, or five or six ways with piano and guitar, at the full sample rate you gave it. Make an acapella, an instrumental or a practice track in minutes.
Uses one minute of your allowance per minute of audio: a four-minute track costs four minutes.
Hear it
Hear it on your own track, free
Free to hear, level-matched against your original. Up to 5 minutes; uploaded, processed, deleted within 24 hours. Keeping the file is a one-time purchase, with no account needed.
Drop a song, or choose one. It is uploaded to our servers, processed, and deleted within 24 hours. Up to 5 minutes.
How it works
What it measures, what it changes
And what stays yours to decide.
- 01
Drop in any song
A finished mix in almost any format, up to an hour long. Separation works on the stereo file you have, so there is nothing to prepare and no session to export.
- 02
Choose how many ways it splits
Two stems for a vocal and an instrumental, four for vocals, drums, bass and the rest, five to add piano, six to add guitar. Ask for the fewest that do the job: every extra stem has to take its sound from somewhere, so a four-way split of a song with no piano comes out cleaner than a five-way one.
- 03
Every stem at full quality
The separation runs at your file's own sample rate instead of squeezing it down to 44.1 kHz and back, so the stems keep their air and their top end. The stereo image is preserved, not rebuilt from a mono guess.
- 04
Keep going in the Studio
Every stem lands in your library as its own file. Send the vocal to Vocal Tuning, rebuild the mix in AI Mixing, or master the instrumental, without downloading anything in between.
The specifics
What you would ask before trusting it with your audio
The formats, standards and numbers behind every result, each one measured.
- Stem counts
- 2 stems (vocals, instrumental), 4 (vocals, drums, bass, other), 5 (adding piano) and 6 (adding guitar). Each layout is its own separation, not one split carved up afterwards.
- Sample rate and depth
- Stems come back at the source sample rate up to 192 kHz, as 24-bit or 32-bit float. No forced downsampling to 44.1 kHz, and no upsampling a lower-rate result to disguise one.
- Nothing goes missing
- The stems add back up to your track within what the model leaves over, and that remainder is returned inside the other stem instead of being thrown away. Nothing in your song quietly disappears between the file you gave and the files you get.
- Built for the hard cases
- Dense low mids, heavy bus compression, reverb tails that belong to the part being pulled out, doubled vocals panned wide: the quality bar is measured on exactly these songs, not on easy ones.
- Ready for the next step
- Every stem lands in your library as its own full-quality file, ready to tune, remix, master or take into your own software.
Questions
What people actually ask
- How many stems should I ask for?
- The fewest that answer your question. For an instrumental or an acapella, ask for two: a four-way split has three more chances to leak. To rebuild a mix, four. Five only if the song really has a piano to find, six only if it has a guitar part worth its own stem.
- Why does the vocal stem have reverb in it?
- Because in the mix the reverb is part of the vocal: a tail printed onto the vocal bus belongs to it. Taking it off is a different job, echo reduction in Audio Restoration, and you can send the stem straight there.
- Can I use the stems commercially?
- That depends on your rights to the recording, not on us. Separating a song you have no rights to does not create rights in the stems. Our acceptable use policy spells this out, and the uploader stands behind the rights.
- How many minutes does separation use?
- One minute of your allowance per minute of song, the base rate. Every service's rate is published, so you know what a run costs before you start it.
- Is this the same thing as a stem splitter?
- Yes. Stem splitter, song splitter, vocal remover and stem separation all name the same job. This is Euphona's version: any song split into music stems at the sample rate you gave it.
Try it
Put your own track through it
The free plan runs the whole engine on your whole recording and plays you the result, matched for loudness against the original. No card. You pay when you want the full-quality file, one track at a time, or with a plan if you make a lot of them.
Who it’s for