Remove Audio from Video
Strip the soundtrack out of a video, with the picture copied rather than compressed again.
- Free
- No sign-up
- No watermark
- Runs in your browser
Loading the tool.
What happens to each part of the file?
A video file is a container holding separate tracks: one for the picture, one for the sound, sometimes several more. Muting never touches the picture. A new container is written, the video track goes into it, and the audio track is simply left behind. Here is what that means for every piece of what you dropped in.
| Part of the file | What happens to it | Quality change |
|---|---|---|
| Video track: H.264 in an MP4 or MOV | Moved into the new file as it is, never decoded | None. The frames are identical |
| Video track: VP9 in a WebM | Moved into the new WebM as it is, never decoded | None. The frames are identical |
| Video track: HEVC, AV1 or VP8 | Decoded and compressed again, as H.264 for MP4 and MOV or VP9 for WebM, at the high quality setting — and HEVC only where the device decodes it | The ordinary cost of one encoding pass |
| Audio track | Left out of the new file completely | Gone, which is the point |
| Frame size, frame rate, length | Untouched, with no menu offered to change them | None |
| The rotation flag a phone writes | Carried into the new file | None. A portrait clip still plays upright |
| Subtitle tracks, chapters, a second language | Not carried over: one picture track is written and that is all | Those tracks are lost |
Whether the picture is copied or compressed again is not a switch you set: it follows two facts about your own file. First the codec, because this tool writes H.264 into MP4 and MOV and VP9 into WebM, so a track already in that codec can be moved across whole. Second the timing: a track whose first frame sits a fraction of a second before 0:00, which is how several export tools hide an encoder's warm-up, cannot be copied from zero and is re-encoded instead. Only the picture's timing counts here, because the sound is being thrown away — and it is the sound that usually starts early, so plenty of files that a format converter would have to re-encode are copied by this one. The page states which of the two it expects before you press anything, then reports what actually happened.
How it works
Drop the video in. Pick an MP4, MOV, WebM or MKV, or drag it onto the box. It opens on your own device, and only the header is read to find the length, the frame size, the codecs and whether there is a soundtrack at all — the same instant's work for a 20-second clip or a two-hour recording.
Read what it plans to do. Before anything runs, the page says in plain words whether your picture will be moved across untouched or compressed again, and which of the two conditions decided it. A file that has no soundtrack to begin with is named as such rather than processed for nothing.
Mute it and download. Press the button and follow the bar; Cancel stops it wherever it is and keeps nothing. When it finishes, the silent video plays right here, next to how much smaller it came out and whether the frames were copied or re-encoded.
What you get
- Nothing is re-encoded when the format allows it
- Taking a soundtrack out is a repackaging job, not a conversion: the already-compressed picture moves into a fresh container and the sound stays behind. When that path is open, the frames you download are bit for bit the ones you started with, and the work takes about as long as reading the file off your disk.
- The file shrinks by exactly the audio
- A repackaged result is smaller by the size of the soundtrack and by almost nothing else. Stereo AAC at 128 kbps costs 0.96 MB a minute, so ten minutes of it weigh 9.6 MB; an uncompressed 48 kHz stereo track, which screen recorders and some cameras write, costs 11.52 MB a minute, and dropping it changes the file dramatically.
- Frame size, frame rate and rotation survive
- There is no resolution menu and no quality slider on this page, on purpose: a 4K clip at 60 frames a second comes back 4K at 60 frames a second, and the flag your phone wrote to say which way up the clip goes travels into the new file with it.
- Long recordings are read in slices
- Your file is never held in memory all at once. It is read a piece at a time as the work moves through it, and the result is assembled the same way, so the limit on a two-hour recording is what your device can hold of the finished file rather than of the source.
Questions people ask
How do I remove the audio from a video?
Drop the file on the box above and press the button. Nothing else is asked of you: no format menu, no quality choice, no sign-up, because muting has exactly one outcome. The silent version appears on the page with a Download button next to it, and the file on your disk is left alone, so you always still have the copy with sound in it.
How do I remove only part of the audio?
You cannot here: this tool takes all of it out and has no way to silence one stretch while keeping another. Two things come close. When the part you want gone sits at the start or the end, cut it away with the video trimmer and everything left keeps its sound. When it is a music bed under someone talking, the vocal remover in the paid product pulls the two apart. Silencing a single word in the middle needs a real editing timeline, and this is not one.
Can I lower the volume instead of removing the sound?
Not on this page. There is one setting and it is off. Turning a soundtrack down rather than out means decoding every sample, scaling it and encoding it again: the picture would still be copied, but the sound would lose a generation of quality passing through, and you would be choosing a level without being able to hear it. Do that in whatever app you edit in, where you can listen while you set it.
Does removing the audio lower the video quality?
Not when the picture is copied, which is the usual outcome for H.264 inside an MP4 or MOV and for VP9 inside a WebM: those compressed frames are moved into the new file without ever being decoded. When the picture does have to be compressed again — an HEVC clip straight off an iPhone, an AV1 download, or a file that stores its video from just before 0:00 — what you pay is the ordinary cost of a single encoding pass at the high quality setting. That path also has to decode the original first, which for HEVC depends on the device rather than on the browser: Apple hardware handles it, and Chrome or Edge do where the computer's graphics hardware supports it. The page says so before you press anything, and the result line tells you which of the two you got, from what the job did rather than from what it expected.
How much smaller will the muted video be?
Smaller by the weight of the soundtrack, which is its bitrate multiplied by its length. The common case, stereo AAC at 128 kbps, works out at 960,000 bytes a minute: roughly 0.96 MB a minute, 9.6 MB over ten minutes, 57.6 MB over an hour. A 256 kbps track is double that. Uncompressed 48 kHz 16-bit stereo, at 11.52 MB a minute, dwarfs both. You are not asked to work it out: the finished panel prints the old size and the new one side by side.
Can I mute a video on my iPhone or Android phone?
Yes, in a mobile browser that has WebCodecs — a current Chrome on Android, or Safari 26 or newer on an iPhone. Open this page there, choose the video from your camera roll, download what comes out. A repackage asks very little of a phone. A clip that has to be compressed again asks a great deal, so keep the tab in front while it runs, and expect a long 4K recording to take its time.
Is the video uploaded to take the sound out?
No. This tab reads the file straight off your disk and writes the new one in its own memory. The only thing it fetches while you work is the tool's own code, never a byte of your recording. That is also why nothing is stamped on the picture, no length or resolution ceiling is imposed and no account is asked for: there is no server bill here for us to recover from you.
Silent on the feed, still understood
Muting is usually deliberate: a music bed you have no licence for, a room full of noise, a feed that starts playing with the sound off anyway. What goes missing with it is the meaning. Keep your original, hand that to GPT-Video, and it transcribes the speech and animates the words over the picture, so the clip still lands for someone scrolling in silence.
Related
Your file is processed on your device and never uploaded.