Free tool

Auto caption generator, every word lit as it is said.

Drop a video and get captions in the style of short-form video: a few words at a time, the word being spoken lit up, popping or boxed. The speech is written out in your browser, you fix any word in the list, pick a font, a colour and a place, crop to 9:16 or 1:1 if you like, and save an MP4 with the captions burned in, or just the SRT.

Drop a video here, or click to choose MOV, MP4 or WebM with someone talking. It stays on your computer; the speech model downloads once.

Your video is read and written in your browser. Nothing is uploaded.

How it works

Captions people actually read.

01 / Listen

Word-timed transcript

Whisper writes out the speech with a start and end time for every word, in your browser.

02 / Style

The word being said, lit

Short captions of one to six words, broken at pauses, with the current word popped, boxed or filled in.

03 / Save

MP4 or SRT

Every frame is drawn again with the captions and saved with the original sound; or take the SRT for any editor.

Questions

Good to know.

How accurate are the captions?

Clear speech comes out well; names, jargon and noisy rooms less so. Open "Fix the words" and correct any caption before you save.

Can I make vertical videos?

Yes: pick 9:16 and the middle of the frame is kept, ready for Shorts, Reels and TikTok.

Does it work in other languages?

Yes, pick the spoken language. Capital letters can be turned off for languages without them.

Is my video uploaded?

No. The speech model and the video work all happen in your browser.