Colurvist › Learn › Add captions to a video automatically
How to add captions to a video automatically
Upload the clip, press Auto in the Captions panel, correct a word or two, choose a style, export. Word-level highlighting, four presets, .srt in and out — in Colurvist's browser editor, for 1 credit a minute of audio.
- Import the clip. Open colurvist.com/editor and upload the video. Captions are generated from its audio track, so the dialogue has to be in the file.
- Open Text → Captions and press Auto. Speech-to-text transcribes the clip and lays each line on a caption track under the video, with a timestamp for every word.
- Fix the words. Click any caption to edit its text. Names and brand words are the usual corrections. Drag a caption's edges to retime it if it lands early or late.
- Pick a style. Four presets: Clean, Boxed, Outline and Karaoke. Karaoke highlights each word as it is spoken, using the word timestamps from step 2. Set size and position — low-centre for landscape, mid-frame for vertical so platform UI does not cover it.
- Export. Press Export. Captions are burned into the picture, wrapped at 86% of the frame width so nothing touches the edges. For platform-native captions instead, use Export .srt and upload the file alongside the video.
Which style for which platform
| Style | Looks like | Use it for |
|---|---|---|
| Clean | Plain white text, soft shadow | YouTube, talking-head, anything landscape |
| Boxed | Text on a dark pill | Busy backgrounds, screen recordings |
| Outline | Bold with a dark stroke | Vertical video over bright footage |
| Karaoke | Each word lights as spoken | TikTok, Reels, Shorts |
Questions
What do auto captions cost?
Transcribing costs 1 credit per minute of audio (a 90-second clip is 2 credits). Correcting, styling, burning captions in and .srt import/export are free.
Can I import captions I already have?
Yes — Text → Captions → Import .srt. They land on the caption track with their timings and can be restyled like any other.
What is word-level highlighting?
Each word lights up as it is said, the style TikTok and Reels made standard. Colurvist uses the transcription's own word timestamps for it, so the highlight is on the beat rather than spread evenly.
Do captions export in the video?
By default they are burned in, so they show on every platform. Export .srt if you would rather the platform render them.
Read next: Veo vs Kling vs Seedance · Colurvist vs CapCut · Colurvist vs Higgsfield · Color grade a video · Remove an object · Auto captions · All models
Last reviewed 2 September 2026. Prices are Colurvist's own and change with the providers' rates.