Bring your track
Choose an MP3, WAV, M4A, or FLAC you own or have permission to process. At launch: up to 10 minutes and 100 MB per song.
Make room for your voice
Keep the band. Take the lead. Explore a backing-track workflow for rehearsals, covers, and your next sing-along.
A workspace for practice, remixes & new ideas.
Drop your audio, or choose a file.
MP3, WAV, M4A, FLAC · Up to 100 MB / 10 min at launch
Preview only · No audio is uploaded
Listen a little closer
An original synthesized arrangement, played as a mix and its source parts. No sung vocals; this illustrates stems, not AI separation quality. Audio source & credits
The planned workflow
Choose an MP3, WAV, M4A, or FLAC you own or have permission to process. At launch: up to 10 minutes and 100 MB per song.
Choose vocals and instrumental, or vocals, drums, bass, and other instruments. Processing and progress tracking are coming next.
The planned free tier offers one short preview per day. Full-track WAV and MP3 downloads will be available with paid minutes.
A closer look
A vocal remover aims to keep the instrumental accompaniment while reducing the lead and backing vocals in a mixed song. That is useful when you want to sing the melody yourself, study the arrangement behind a vocalist, or rehearse an entrance without following the recorded performance. The result depends on how the voice interacts with the instruments, particularly when guitars, synths, and harmonies share a similar range.
For a backing track, first listen to the chorus at the volume you expect to use during rehearsal. Check whether harmonies remain, whether the drums retain their impact, and whether the chords sound steady. Then check a quiet verse where small traces of the original singer may be easier to notice. The best result for singing practice is a stable accompaniment with clear timing, even if a little reverb remains in the background.
Removing vocals can also change sounds that resemble a voice, such as a sustained lead instrument or choir-like synthesizer. Conversely, effects layered around a vocal can remain in the instrumental output. If you need the singer on their own instead, use a vocal isolation workflow. If you want to practice a full karaoke performance, pay attention to the intro length, instrumental breaks, and ending as well as the vocal-free sections.
Keep a copy of the original recording, and avoid adding compression or effects before separation. Turning a low-quality file into a larger format will not restore detail. Use the same section when comparing outputs, and listen at similar volume. The most useful result is the one that serves your next step, whether that is learning a phrase, rehearsing with a backing track, or testing an arrangement in your audio editor.
Good questions
It can. Backing harmonies, vocal reverb, and heavily processed singing may leave remnants. A real preview is the best way to judge whether they affect your use.
Some musical sounds overlap with the voice. A separation model can soften those sounds or leave audible changes in the accompaniment.
Both can come from a two-stem split. Removal focuses on the instrumental output; isolation focuses on the vocal output.
Processing is not active yet. This page offers a file-selection preview and original sample stems to explain the workflow.
You need the appropriate rights for the recording and your intended distribution. Separation itself does not grant permission to publish or monetize a backing track.
Keep exploring