Unmix · StemStudio — free online stem separation
Extract the vocal part from a song to create an acapella track.
Free in your browser · instant
Your audio is never sent anywhere. Everything is processed inside your browser.
Highlights
Separate vocals with an htdemucs-family model and export them directly.
It doesn’t rely on frequency filtering alone, so the result tends to sound natural.
Your audio is processed only inside your browser.
Download the extracted vocals directly.
When to use it
Three simple steps
Drag & drop an MP3, WAV, M4A, FLAC or video, or click to select.
Instant acapella extractor. AI mode can split into 6 parts.
Preview and adjust each part, then download what you need as WAV.
Accuracy is fairly high thanks to AI-model separation. It automatically picks a high-quality or lightweight model by device — both genuine AI. For a finer split by instrument, you can also use AI 6-stem.
Remixes, practicing covers, checking your ear-transcription, and more.
It’s completely free with no account needed. Just open the page and pick a file.
No. All separation happens entirely inside your browser.
MP3, WAV, M4A, AAC, FLAC and OGG, plus the audio track of video files.
The file you pick is processed right inside your browser. Nothing is ever sent to a server.
No account, no install. Just open the page and choose a file.
Remove vocals to make a backing track, or isolate just the vocals. Great prep for practice or production.
Try each part’s volume and mute right there, then save only what you need as WAV.
It means automatically splitting one song into six parts. Here is what each one is:
Drag & drop an MP3, WAV, M4A, FLAC or video, or click to select.
If vocals + backing track is enough, choose "Vocal extraction"; to split by instrument, choose "AI 6-stem".
Play and adjust each part, then save just what you need as WAV.
It’s completely free with no account needed. Just open the page and pick a file.
No. All separation happens entirely inside your browser.
MP3, WAV, M4A, AAC, FLAC and OGG, plus the audio track of video files.
Both use the same family of AI model (htdemucs); they differ in how many parts you get. Vocal extraction gives two (vocals and backing track); AI 6-stem gives six (vocals, drums, bass, guitar, piano and other). If you don’t need a specific instrument, vocal extraction is plenty.
Vocal extraction automatically picks between two AI models by device (a high-quality model on PC, a lighter one on phones). Both are genuine AI separation — we never use a low-quality "simple" method. In the rare case neither can load, we ask you to retry on a good connection or a PC.
Please only use audio you own the rights to, or that you are otherwise permitted to use.