Your files never leave your device — 100% local processing

Remove Silence from Video

Finds the parts where nobody is speaking and cuts them out. How much shorter the video gets depends on how many pauses it has; by default only pauses longer than 0.8 seconds are cut.

Cuts leave a little room on both sides so the first sound of a word is never clipped. Short breaths are kept, and every cut is joined smoothly. You can tune the padding and the sensitivity below, and the bar shows you exactly what will be removed.

0 B uploaded — this page has not sent your files to any server?

Recordings of lectures and meetings are awkward files to upload — this tool never uploads them. Listening to the audio, cutting it and saving the result all happen on your own computer.How this differs from upload-based sites — most editing sites upload your video to a server, process it there, and promise to delete it after a while. Until that happens, your file is sitting on someone else's server. vidmoa has no upload step at all. The more sensitive the recording, the more that difference matters.

Remove the silent gaps but keep every word: which setting does what?

Lectures, meetings and walkthroughs are full of pauses while someone thinks or switches slides. This tool measures how loud the audio is, finds the stretches without speech, and cuts both picture and sound there to build a tighter MP4. Before you save anything, a bar shows exactly what will be removed.

A closer look

Detection is based on loudness, not speech recognition. The audio is converted to 16 kHz mono, its level is measured every 10 ms, and the boundary between silence and sound is set from that video’s own quiet end (the bottom 10%) and speech level (the top 5%). Because there is no fixed decibel value, quietly and loudly recorded files both work. Sensitivity runs from 1 to 10: lower removes more, higher is more careful. Keep in mind that non-speech sounds such as typing or laughter count as sound and are kept.

With the defaults, here is what actually gets cut. Each stretch of sound gets 0.20 s of room before it and 0.40 s after it, and a gap is removed only if at least 0.80 s is still left. In other words, the pause between two phrases must be longer than roughly 1.4 seconds, and the amount removed is that pause minus 0.6 seconds. Cuts are placed on frame boundaries and the audio is cut at the same point, with a short fade (15 ms by default) at each join to prevent clicks and a fresh keyframe so the picture doesn’t smear. The audio is re-encoded in the process.

How to use it

When it helps

Recorded lecturesStrip the pauses while you write on the board or change slides, so students get the same content in less time.
Video-call recordingsShorten the dead air between one speaker finishing and the next one starting, for a record that’s easier to rewatch.
Product demos and tutorialsClose the gaps where you stopped talking to operate the software. Silent on-screen actions are cut too, so check that no step you meant to show has disappeared.
First pass on talking-head videosRemove the long pauses up front so there’s less to trim by hand in your editor.

Common snags

“No silence worth removing was found.”The pauses are shorter than the default threshold (about 1.4 s), or background noise keeps the quiet parts from reading as silence. Less room before and after, or lower sensitivity, will catch more.
The start or end of words sounds clipped.Increase Room before and Room after a little. Speech is only recognized once the sound has started, so detection is always slightly late, and short, sharp consonants such as p, t and k suffer most.
“In this recording, silence and speech are close in level.”This appears when fan or air-conditioning noise, or a distant microphone, leaves less than 8 dB between the quiet parts and speech. Results can be off, so move the sensitivity one step at a time, compare the bar each time, and check the result carefully.
No sound in the result, or it’s slow.Putting the trimmed audio back requires the browser’s audio encoder; where that’s missing, you’ll be told the video will be saved without sound, so use a recent Chrome or Edge. The whole video is re-encoded, so time grows with length, and hiding the tab slows it down a lot.

Tip Check the bar with the defaults first; if the result sounds rushed, raise “Only remove gaps longer than” to a second or more, and if too little is cut, drop Sensitivity to 3 or below.

Frequently asked questions

Will the first sound of a word get cut off?

That is exactly why there is room on both sides. Speech can only be recognised after the sound has started, so the decision is always a little late. We rewind by that much (0.20s by default) before cutting, and leave more room at the end (0.40s by default) for the tail of the word and the room reverb. If words still get clipped, increase the room before.

The result sounds choppy. What now?

Raise "only remove gaps longer than". The default is 0.80 seconds. Removing every short breath makes speech sound like a machine gun. Cuts that end up very close together are merged automatically so the result does not hiccup every tenth of a second.

What does sensitivity do?

It decides how quiet counts as silence. Recording levels differ from file to file, so the threshold is not a fixed number — it is measured from the quiet parts and the speech in your own video. Lower sensitivity removes more, higher sensitivity is more careful.

Will the audio drift out of sync?

Every cut is placed on a frame boundary, and the audio is cut at the same point. A very short fade is applied at each join so there is no click.

Which files can I use?

MP4 and MOV, iPhone recordings included. A video with no sound cannot be processed, because there is no silence to find. For other formats, convert first — for example WebM → MP4.

Is my file uploaded?

No. Everything runs inside your browser, and your file is never uploaded. The page does need a connection to download its processing scripts when you first use it.

Is it free?

Completely free. No limits, no watermark, no sign-up.