Remove Background Music from Video (Keep the Voice)
AI separates the voice from the music in your video, on your own device, and gives you the video with the voice only. Nothing is uploaded, and there's no signup or watermark.
Your file never leaves your device. How?
Your media is processed in your browser and is not uploaded to our servers for conversion. Page assets (including the converter engine) still load over the network.
Select or drop a video
MP4, MOV, WebM or a sound file. The music is removed on your device: nothing is uploaded.
Click, tap, or drop a fileHow to remove the music from a video
Choose the video
Drop or select an MP4, MOV, WebM or MKV video, or a sound file, from your device.
Let the model download once
The first time, about 43 MB downloads (the AI model and its runtime) and is saved in your browser for next time.
Remove the music
The voice is separated from the music on your device, then put back in your video. The picture is copied as it is, so it doesn't lose quality.
Play and download
Play the video to hear the voice on its own, then download it. The voice and the music are also there as MP3s.
What people use it for
Creators remove a song from a clip they filmed so they can use their own voice-over or a track they have the rights to. Teachers and presenters clean the background music out of a recording so the speech is clear. People who make dubs, subtitles or transcripts strip the music first so every word is easy to hear.
Most music removers upload your video to a server, add a watermark or ask for an account. This one separates the sound on your own device, so private and unreleased videos stay private.
How the AI separation works
The sound is read from your video and separated by an MDX-Net model from the open-source Ultimate Vocal Remover project, running in your browser with ONNX Runtime (on your graphics card where the browser allows, otherwise the processor). The model predicts which parts of the sound are voice; everything else is the music. Long videos are done in parts, crossfaded, so phones don't run out of memory.
The voice is then put back into your video with FFmpeg: the picture is copied without re-encoding, and the new sound is encoded as AAC (Opus for WebM). Speech over music separates well; singing, heavy effects or a voice that sits inside loud music can leave a little behind, and sound effects go with the music.
Features
Never uploaded
The video stays on your device: the AI runs in your browser.
Picture untouched
Only the sound changes; the video is copied as it is, with no quality loss.
Voice and music apart
Download the video without music, the voice as an MP3, or the music on its own.
Music Mixer (Pro)
Turn the music down instead of off, let it dip whenever someone speaks, or put your own track under the voice. Hear it live as the video plays; your first mix is free.
Frequently asked questions
Is my video uploaded?
No. Only the AI model is downloaded, once, from this site. The sound is separated in your browser and the new video is made on your device.
How long does it take?
About as long as the video on a laptop, longer on a phone; browsers with WebGPU (recent Chrome and Edge) are faster. The first run also downloads about 43 MB, which is then cached. Keep the tab open while it runs.
Will it remove sound effects too?
Yes: it keeps the voice and treats everything else (music, effects, background noise) as the music. Speech comes out cleanest; singing is treated as voice.
Can I turn the music down instead of removing it?
Yes, with the Music Mixer (Pro): set the voice and music levels, let the music dip automatically whenever someone speaks, or swap in your own track. You can try every setting and hear it on your video for free; your first mixed video is free too.
Does removing the music fix a copyright claim?
Not by itself. A platform may still match what's left, and a claim can cover more than the music. Use it to make a clean voice track or to swap in music you have the rights to, and use the platform's own process for an existing claim.
Which files and lengths work?
MP4, MOV, WebM, MKV and most videos, or a sound file. Up to 15 minutes on a computer, or 6 minutes on a phone, because the sound is separated in memory.
Popular conversions
Related: Vocal Remover · Noise Remover · Remove Audio from Video · Auto Subtitles · All tools