Why the order matters
Every cut moves the words after it earlier. If captions are made first and the fillers removed afterwards, each caption after the first cut is late unless the tool re-times it. OpusClip’s API guide says captions added before cleanup shift and have to be redone. Doing the cleanup first and the captions second avoids that, but it means two passes.
How it works in Shorty
Shorty transcribes each clip once, word for word, and shares that transcript between filler removal and captions. When the director cuts the fillers, it rebuilds the caption timing, sound cues and the edit’s length from the cut transcript. The captions you see are the words as spoken in the final video, with nothing to re-sync.
How other editors handle it
| Tool | Filler removal | Captions |
|---|---|---|
| Descript | Remove filler words in AI Tools | Captions follow the edited script |
| CapCut | Identify filler words while generating auto captions | Transcript edits are linked to captions |
| VEED | Magic Cut | Subtitles generated in the same editor |
| OpusClip | Speech Cleanup | Captions added before cleanup shift and need redoing, per its API guide |
Sources: Descript: filler words · CapCut: filler word removal · VEED: Magic Cut · OpusClip: filler-word removal API guide. Competitor features checked 20 October 2026 in each tool’s own documentation. Plans and features change, so check the current page before you choose.
Example brief
Adapt this brief to your footage and review the captions against the speech.
Cut this interview into a 40-second vertical video. Remove the ums, uhs and false starts, open on my strongest sentence, and add bold word-by-word captions at the bottom.
See it on four real talks
We ran four public talks through the same tool the director uses inside an edit. Across them, 107 words came out in 80 cuts, including every “um” and “uh” in the transcripts, and every kept word survived in order. The launch post plays each one before and after, with every removed word struck through.
| Recording | Length before → after | Words cut | Cuts | Ums and uhs left |
|---|---|---|---|---|
| Nik Storonsky, Revolut | 2:00 → 1:49.8 | 40 | 28 | 0 of 13 |
| Prime Intellect, NVIDIA event | 1:00 → 0:53.2 | 38 | 27 | 0 of 16 |
| Arjun Karanam, Trajectory | 2:00 → 1:54.4 | 22 | 21 | 0 of 8 |
| Elon Musk, Tesla Roadster event | 2:00 → 1:58.4 | 7 | 4 | 0 of 1 |
Sources: Shorty News: four talks before and after. Measured on excerpts of public YouTube talks on 20 October 2026. Results depend on the recording.