AUTO CAPTIONS

How to use an auto caption generator for your video

AI Free Audio to Video · Last updated 2026-10-20 · 5 min read

THE SHORT ANSWER

An auto caption generator transcribes the speech in a video into timed text. In Edit subtitles, upload your own video, review the Whisper draft, choose a caption style and export. MP4 puts captions into the picture; SRT keeps them in a separate text file. Google sign-in is required. The current service is free, with no site watermark, for files up to 100 MB and 10 minutes.

Choose video captions or a social post caption

Here, caption means spoken words shown at the right moments in a video. The tool does not write promotional post descriptions, hashtags or image captions. Use Edit subtitles when the original footage should remain visible. Use Subtitle generator when you only need an SRT file.

Upload the recording and select its spoken language

Upload MP4, MOV, WebM or MKV with sound. Choose English, Chinese, Japanese, Korean, Spanish, French or German, or automatic detection. Choose the language being spoken, not a target translation language. The server runs Whisper to produce a draft. Platform URLs and live recording are not accepted.

Correct the transcript before adding animation

Listen to names, numbers, quiet speech and sentence endings. Fix incorrect words and check each line’s start and end. Recognition is an estimate, especially with singing or overlapping voices. Replacing a phrase estimates the timing of its new words; play that section again before exporting.

Style readable word-by-word captions

Compare pulse, marker or neon for animated emphasis, and minimal, editorial, outline, mono or typewriter for other treatments. Adjust word groups, size, position and colors. Keep captions away from faces and important details. A short group can be easier to read than a complete sentence on a narrow phone screen.

Export MP4 for visible captions or SRT for editable timing

MP4 contains the rendered captions and their visual style. SRT contains only text and cue timing, so it cannot carry fonts, colors or animations. Play the exported MP4 with sound, including any cuts, zoom, B-roll and music you enabled. Private uploads and completed results expire after 14 days; shared queue and hourly limits apply.

Common questions

Can I generate free captions without a watermark?

Yes. Current MP4 exports do not add a site watermark. Google sign-in is required, with 100 MB and 10-minute file limits and shared usage limits. This does not remove a watermark already present in your source.

Does an AI caption generator guarantee accurate subtitles?

No. Check the draft against the recording and correct words and timing before publishing. The public examples show styles, not an accuracy guarantee.

Ready to give it a try?

Start with one file. Make something new.

Generate captions on your video