Mute Part of a Video

Silence one part of a video, keep the rest

Pick a video, type the start and end of the moment you want to hush, and get the same clip back with the audio silenced only between those two times — everything before and after keeps its original sound. It's the surgical version of muting: perfect for killing a swear word, a copyright-flagged song in the background, a burst of mic noise, or an off-hand comment, without wiping the whole soundtrack.

The picture is copied untouched — the video stream is stream-copied losslessly (-c:v copy), so there is no quality loss and no re-render of the image. Only the audio is re-encoded, because the samples inside your window are being zeroed. Everything runs with ffmpeg inside your browser tab, so nothing is uploaded, and the tool is free with no watermark.

How the times work

Both Mute from and Mute until are seconds from the start of the clip. They can be whole numbers or decimals:

• 5 and 10 silence the sixth through tenth second.
• 0 and 3 silence the first three seconds.
• 12.5 and 13.2 silence a fraction-of-a-second window — enough to drop a single spoken word.

Mute until must be greater than Mute from; the range in between is silenced and the audio outside it is passed through unchanged.

Worked example

You recorded a product demo and someone says a name you'd rather not publish at around the 12-second mark. Load demo.mp4, set Mute from to 12 and Mute until to 13.5, and run it. You get demo-muted.mp4 — the same video, same picture, with just that 1.5-second window silenced and the rest of the narration intact.

On the command line the same job is gizza tool video-mute-section url=https://example.com/demo.mp4 start=12 end=13.5 --out demo-muted.mp4, and the page URL /tools/video-mute-section/?start=12&end=13.5 deep-links to the same pre-filled settings.

Notes and limits

• One range per run. To silence several separate moments, run the tool once per range (feed the output of one run into the next), or pick a single window that covers them all.
• The whole audio track is silenced in that window — this mutes everything (voice, music, ambience) between the two times; it can't isolate one voice or one instrument (that needs an AI model).
• The video is copied losslessly; only the audio is re-encoded. mp4, mov, m4v and mkv keep their container and use AAC audio; webm keeps webm and uses Opus; other inputs come out as MP4.
• If the video has no audio track there is nothing to silence and ffmpeg will report an error — use a clip that actually has sound.
• Input and output are each capped at 25 MB (the file is processed in your browser's memory).