How to cut pauses without clipping words
Automatic silence detection is a starting point. Audio varies by microphone, room, and speaker, so reviewing every boundary is what protects the beginning and end of each phrase.
1. Prepare your clips
Use MP4 or MOV files with audio. The tool accepts two to six clips, up to 200 MB per file, 500 MB total, and 15 minutes combined. Put them in final order before analysis. Your original files remain unchanged.
2. Tune detection for the recording
The analysis measures audio level; it does not recognize words. If speech is quiet, move sensitivity to a more negative value to preserve softer sounds. If steady background noise is present, a less negative value may help separate pauses, but it can also treat quiet syllables as silence.
Set the minimum pause duration you want to shorten. Pauses below that threshold remain unchanged. For longer pauses, the tool keeps a margin before and after detected audio; increase those margins when words begin or end softly.
3. Preview both sides of every cut
Preview each clip, especially the beginning and end of kept regions. Check that the first consonant and final syllable remain audible. Music, breaths, echo, or room noise can confuse detection. On mobile, the preview stays pinned at the bottom while you scroll through the list.
4. Remove whole detected segments when needed
If a detected segment should not appear in the finished video, use “Cut this video segment.” It removes the segment from the edit; adjusting neighboring boundaries will not put it back on the timeline. Then review the transitions to avoid an abrupt start.
5. Export and review the MP4
Choose a hard cut, dissolve, or fade through black. Export creates an H.264 MP4 with AAC audio and re-encodes the material; file size and quality can change. Play the downloaded file from start to finish before publishing or replacing any original.
Privacy and limits
Analysis and assembly happen in your browser. FFmpeg WebAssembly code loads on demand from the CDN listed on the tool page; selected clips are not sent to our server. Large videos require memory and may slow down phones.