Remove Silence From Video Automatically (Auto Cut)
AIRemove silence from video automatically, and the “um”s and “uh”s with it. Auto Cut measures where the sound drops below a set level, finds filler words with on-device speech recognition, and joins what is left in one pass with click-free cuts. You choose how long a pause must be and how much breathing room stays. Nothing is uploaded.
- Runs on your device
- Whisper AI
- Web Audio
- Works offline
Updated
Drop your video or audio file here
How to remove silence from a video
Step 1: Drop a video or a recording on the box above.
Step 2: Keep Cut silences and Cut filler words on, or switch one off. Set Shortest pause to cut and Keep around speech.
Step 3: Press Cut. With filler words on, the speech model downloads once after you confirm.
Step 4: Watch the result: the card shows how many seconds and pauses were removed, and which fillers. Then download it.
Good to know
Settings explained
- Cut filler words (um, uh…)
- Uses speech recognition with word timings.
- Spoken language
- Picking the language skips detection and improves accuracy on short clips.
- Extra filler words
- Your own verbal tics, separated by commas. Cut wherever they are spoken.
- Quality
- Only used when a stream has to be re-encoded. Compatible streams are copied with no quality loss.
- Faster: Smaller file, quickest export
- Balanced: Good quality at a sensible size
- Best quality: Largest file, slowest export
Formats
How it runs
- Whisper speech model
- Speech recognition uses an open Whisper model. It downloads once after you confirm its size, then runs on your device — on the GPU via WebGPU when available.
- Web Audio
- Audio is processed with the browser’s built-in Web Audio engine, sample-accurate and fast.
- Hardware video engine
- Video is decoded and encoded by your device’s media hardware through WebCodecs — usually several times faster than real time.
Auto Cut settings explained
The defaults suit a talking-head video or podcast: pauses of 0.7 seconds or more are removed, and 0.15 seconds of room is kept before and after speech so words never sound clipped. Lower the pause length for a faster, punchier edit; raise the breathing room for a calmer one.
| Setting | Range (default) | What it changes |
|---|---|---|
| Cut silences | On / off (on) | Removes stretches where the sound stays below the silence level |
| Cut filler words | On / off (on) | Finds “um”, “uh” and similar with word-timed speech recognition |
| Shortest pause to cut | 0.2–3 s (0.7 s) | Shorter pauses are left alone |
| Keep around speech | 0–0.5 s (0.15 s) | Room left before and after each kept phrase |
| Silence below (Advanced) | −60 to −20 dB (−42 dB) | Raise it for noisy rooms, lower it for very quiet speakers |
| Extra filler words (Advanced) | Your own list | Cuts your verbal tics, such as “basically” or “you know” |
Which filler words are removed?
Hesitation sounds are always cut. Words that are fillers only sometimes, such as “like”, are cut only when commas or a short pause set them apart, so “I like this” stays intact. A cut never reaches into a neighbouring word.
| Language | Always cut | Cut only when set off |
|---|---|---|
| English | um, uh, er, erm, ah, hmm | like, basically, literally, you know, I mean, kind of, sort of |
| Spanish | eh, em, ehm | bueno, pues, tipo, este |
| French | euh, heu, bah, ben, hum | genre, bon, alors, quoi, voilà |
| German | äh, ähm, öh, hm | also, halt, sozusagen, quasi |
What is a jump cut?
A jump cut is a cut between two moments of the same shot, so the speaker seems to jump slightly. YouTubers use them on purpose to remove dead air and keep the pace up, and Auto Cut works as an automatic jump cut editor that makes them for you. The video is re-encoded once with the gaps dropped, and the sound is trimmed sample by sample with a few milliseconds of fade at every join, so cuts do not click.
Auto cut silence without Premiere Pro or CapCut
- No editor to install or learn. Remove pauses from video in one pass: drop the file, press Cut, download.
- Private. Speech recognition runs in your browser; the recording never leaves your device.
- Free and unwatermarked, with no export limit.
- Audio too. Podcasts and voice recordings come out as audio; MP3, WAV, FLAC and M4A keep their format.
Tips & troubleshooting
Audio-only recording with no filler words to worry about? Remove silence from audio is quicker and has no model download.
Background hum keeps “silence” above the threshold. Clean it with remove background noise first, or raise Silence below.
Cutting before captioning gives tighter captions: run the result through the subtitle generator.
For a cut you want to place by hand, trim the result with the video trimmer.
Auto Cut (Remove Silences & Filler Words) FAQ
How do I remove silence from a video automatically?
Drop the video here, keep Cut silences on and press Cut. Every pause longer than 0.7 seconds (adjustable from 0.2 to 3) is removed, with 0.15 seconds of room kept around speech, and the remaining parts are joined into one video.
How do I remove filler words from a video?
Keep Cut filler words on. A speech model finds each word with its timing, and hesitations such as “um” and “uh” are cut out precisely. Add your own habits, such as “basically”, under Extra filler words in Advanced settings.
Will the cuts look jumpy?
They are jump cuts, so the picture moves slightly at each join; that is the accepted style for talking-head videos. Keeping 0.1–0.2 seconds around speech makes them feel natural, and the audio fades over a few milliseconds at every cut so there are no clicks.
Why was an “um” left in?
Speech recognition sometimes treats a hesitation as noise and does not write it down, so it cannot be found as a word. If the hesitation is followed by a pause, the pause is still cut, and a sound written with an unusual spelling can be added to Extra filler words.