Reasons to strip the sound
A silent copy of a video is useful more often than it sounds:
- Privacy. Phone clips pick up conversations, names, addresses read aloud and background TV. Removing the track before posting a clip of a flat, a product or a child's first steps avoids sharing more than the picture.
- Website backgrounds and product loops. Browsers only autoplay video that is muted. A file with no audio track at all is also slightly smaller and behaves predictably in every player.
- Wind and noise. Outdoor footage, drone clips and dashcam recordings often carry nothing but wind roar or engine drone.
- Replacing the soundtrack. Editors and social apps make it easier to add music to a clip that has no competing audio. Starting silent avoids accidental mixing.
- Copyright. A clip with a song playing in the background can be blocked or demonetised on video platforms even if the picture is entirely yours.
- Smaller files. On long, mostly static recordings, audio can be a meaningful share of the size.
What happens to the file
For H.264 and HEVC video — what phones, cameras, OBS and most downloads produce — the tool does not re-encode anything. FFmpeg reads the container, leaves out every audio stream and copies the video stream into a new MP4 exactly as it was compressed. In practice this means:
- No quality change. The pixels in the output are the same bits as in the input.
- Speed. The work is reading and writing the file, not compressing it, so a 1 GB clip is usually done in well under a minute even in the browser.
- Resolution, frame rate and codec stay the same. A 4K 60 fps HEVC clip from an iPhone remains 4K 60 fps HEVC.
- Orientation is kept. The rotation flag that phones store for portrait video is copied along with the stream, so the clip shows the right way up.
- Fast start. The MP4 index is written at the beginning of the file, which helps when the clip is used on a web page or sent through a messenger.
The output size is the original size minus the audio. For typical phone video with 128 to 256 kbit/s AAC sound, expect a saving of 1 to 3%; for a long screen recording with little motion, the saving can be much larger.
Because only the audio is removed, this is also the safest first step before other edits. Muting a clip, then using compress video on the silent version, gives a smaller file than compressing first, since no bits are spent on sound you will discard.
Which inputs work
MP4, MOV, WebM and MKV files are accepted, and the result is always an MP4. The tool first checks the video codec:
- H.264 and HEVC are copied losslessly, as described above. An HEVC clip stays HEVC: it plays wherever the original did, and muting does not make it more compatible.
- Everything else — VP8 and VP9 from WebM files and browser screen recorders, AV1, ProRes from Final Cut exports and professional cameras — is re-encoded to H.264 at CRF 23, because MP4 players cannot rely on those codecs. This takes roughly as long as the clip on a laptop instead of seconds, and the quality change at CRF 23 is hard to see.
Browser or server
By default everything runs in your browser with FFmpeg compiled to WebAssembly, and the video never leaves your device. Files up to 2 GB are accepted; on phones, very large clips can exhaust browser memory because the input and output are held in memory together.
The page also offers server processing with the same approach using native FFmpeg. It avoids browser memory limits and suits multi-gigabyte files. The free plan includes one server job a day up to 500 MB, and Premium removes the daily limit and accepts files up to 4 GB. Server files are deleted automatically within about an hour.
Related edits
- To keep the sound and drop the picture instead, extract the audio with MP4 to MP3.
- Several files can be muted in one go; they are processed one after another and can be downloaded individually or as a single ZIP.