How to remove “um”, “uh” and long pauses from a video automatically
· 4 min read
Ninety seconds of talking usually holds seventy seconds of speech. The rest is pauses and “um”s, and they are what make a talking-head video feel slow. Cutting them by hand means scrubbing for every one. Auto-cut does it from the transcript in one step — and it is careful about which words it treats as filler.
Steps
Open your video and generate captions. Filler words are found from the word timings speech-to-text produces, so captions come first.
Open Auto-cut. Choose whether to remove silences, filler words, or both, and how long a pause must be to go: 0.4, 0.6 (the default), 1.0 or 1.5 seconds.
Check what it found. It reports how many silences and fillers it found and how many seconds that saves, before changing anything.
Apply, then undo if you disagree. One press removes them all; one undo puts them all back. Captions and sound effects move with the footage.
Which words count as filler
Only sounds that are almost never meaningful. In English: um, uh, erm, hmm, ah, er, eh and their spellings. In Hinglish and Hindi: matlab, yaar, accha, haan — which do carry meaning sometimes, so check those cuts. Spanish, French, German and Portuguese have their own lists.
“Like” and “so” are not on the list, on purpose. They are real words far more often than fillers, and cutting them mid-sentence would break the meaning. Remove those by hand when they are filler.
Picking the pause length
0.6 seconds suits most talking-head videos: it removes the thinking pauses and keeps the natural gap between sentences. 0.4 gives the fast, jump-cut pace of many Reels — use it with captions, because the cuts are frequent. 1.0 or 1.5 is for interviews and podcasts where a pause carries meaning.
Tight cuts make a static camera jump between frames. A slight zoom on alternate cuts hides it — the AI video editor does that for you.
Questions
How do I remove filler words from a video automatically?
Generate captions in AI Caption Cut, open Auto-cut, tick “Remove filler words”, and apply. It removes um, uh and similar sounds using the word timings, and keeps the captions in sync.
Can it remove silences too?
Yes. Auto-cut removes pauses longer than a length you choose — 0.4, 0.6, 1.0 or 1.5 seconds — along with or instead of filler words.
Does removing filler words break the captions?
No. The captions are rebuilt from the same word timings, so they stay matched to the speech after the cut.
Cuts, captions and pace without editing skills: Edit a video without skills
Keep reading
All articles →How speech-to-text runs inside a browser tab, with no upload
AI Caption Cut transcribes with OpenAI Whisper base, running in the browser through WebAssembly — the video is never uploaded.
10 November 2026 · 7 min readCan ChatGPT or Claude edit videos?
On their own, ChatGPT and Claude cannot cut, caption or export a video file — they write text and code.
10 November 2026 · 5 min readWhy your exported video plays as a black screen (and the sound still works)
A black picture with working sound almost always means the player refused the file before the first frame — not that the file is broken.
10 November 2026 · 6 min read
Related free tools
Try it on your own video
Nothing to install, no account, and your video never leaves your device. Unlimited free exports, and a one-off payment when you want the watermark gone.
Caption my video — freeUpdated