Timed twice
The whole video is transcribed once, then each selected passage is re-transcribed word by word; if the second pass cannot align, the first timestamps are kept and flagged as estimated.
SHORTS CAPTION GENERATOR
Generate word-timed captions on your shorts.
The captions here are generated on the shorts the shortshort studio cuts from your long video, in the language spoken, with each word placed on the moment it is said. It is part of the one shortshort product, not a standalone subtitling service, and it does not caption a whole video you bring in for that purpose alone.
60 free credits, no cardUp to 3 hours per videoMP4 · 1080 × 1920
Four styles, each with a different job
Spotlight highlights the spoken word in uppercase, Impact sets bold uppercase lines with one highlighted word in each, Karaoke lights words up as they are said, Minimal stays quiet with no effect.
Your own SRT is accepted
If you already have an SRT timed to the source video, import it and its cues replace the studio's transcription in the captions of the short.
How to do it
Four steps. About two minutes of your time.
- 01Upload the long video the shorts will be cut from, MP4, MOV or WebM, up to 3 hours
- 02Let the AI pick the passages, or set clip bounds yourself in the free editor
- 03Choose one of the four caption styles and set colour and font for the video
- 04Check the timing on each short, then export the MP4 with the captions rendered on it
What this page does not promise
No standalone captioning: captions are made on the 15 to 90 second shorts cut from your video, not on a full-length file uploaded only to be subtitled.
No translation and no dubbing: a short recorded in French comes out captioned in French, an English one in English, with no second-language track and no synthetic voice.
Questions people ask
Which languages do the captions support?
The captions are written in the language the speaker uses in the video, as transcribed. They are not translated into another language, and there is no dubbed audio track.
How accurate is the timing?
Each word of a selected passage is timed in a dedicated second pass on the audio. When that pass cannot line up with the first transcription, the original timestamps are kept and marked as estimated, so you know which captions to check in the studio.
Can I use captions I already have?
Yes. Import an SRT file timed to the source video and the studio uses those cues on the shorts instead of its own transcription.
Can captions be styled to match my brand?
Yes. Every plan lets you pick the caption colour and the font for each video before rendering. Studio goes one step further and stores that choice as the account's brand preset, a single one per account, applied by default to every new video.
Can I caption a video without cutting it into shorts?
No. The captions are a step of the shortshort studio and apply to the shorts it produces, which run 15 to 90 seconds each. A one-hour video uploaded for captions alone would still come out as shorts.
Which style should I pick for a talking-head short?
Spotlight is the default: uppercase text with the word being said highlighted in turn, which follows the rhythm of speech. Impact suits a clip built around one number or claim; Karaoke suits explanations; Minimal suits clips where the picture matters more than the words.
A 31-minute interview on Creator, 6 shorts, one drop.
Free to try with 60 credits. Your source video stays yours; the shorts you export join the community, and a plan keeps them out of it.