Loading the page…
AI listens to the recording and splits it into vocals and music. Video works too: we extract the sound for you.
Upload audio or video and get two files in about a minute: vocals without music and music without vocals (a backing track). Works with songs, podcasts, phone recordings and social clips. We extract the sound from video ourselves.
Yes. Upload an MP4, MOV or WebM clip, we extract its sound and split it into vocals and music. You can then put the clean voice back on the video in any editor.
MP3, WAV, M4A and AAC audio up to 50 MB, or video up to 200 MB. Up to 10 minutes long.
Two MP3s: vocals on their own and music without vocals. With the ElevenLabs model you can get 6 tracks: vocals, drums, bass, guitar, keys and the rest. Everything appears in your works feed, and in Telegram if you signed in with it.
Songs with studio vocals split very cleanly. Live recordings with noise may keep a little background in the vocals and traces of voice in the music.
Suno gives two versions per run at the same price for any track length. ElevenLabs is faster, returns one version and prices by track length, but can split into 6 tracks and extend by an exact number of seconds. The exact price shows on the button before you run.