Automatically generate subtitles (SRT or VTT) from a video's audio track - locally in the browser via AI speech recognition (Whisper), no upload.
Running locally on your device ...
Running locally on your device ...
0%
Your files never left your device
No. Everything runs in your browser - your file never leaves your device. How this is verifiable
A talk, an explainer video, a social clip without captions: especially on a muted feed or for viewers with a hearing impairment, subtitles often decide whether a video gets understood at all. This tool listens to a video's audio track, recognises the spoken words via AI (Whisper) and writes them out as a time-stamped subtitle file - ready to load into a video player or editing timeline.
The audio track is automatically pulled out of the video - no separate detour through a dedicated extraction tool is needed, the same mechanism the site's audio-extraction tools already use handles that. Whisper breaks recognition into speech segments with a start and end time; those come out directly as SRT or VTT cues, in exactly the same timestamp formats as this site's other subtitle tools (so, for example, still fine-tunable afterwards via "Shift subtitles").
Subtitles are an automatic approximation, not a substitute for an edited translation - proper names, jargon and background noise can cause errors. A quick review pass before publishing is always worthwhile.
| Input formats | MP4, M4V, WEBM, MOV, MKV |
|---|---|
| Auto-conversion | PNG, GIF |
| Output formats | SRT, VTT |
| Batch processing | No |
| Processing | Locally in your browser (WebAssembly) |
| File upload | None |
Limitations: Runs single-threaded (no cross-origin-isolation mode on this site) and is therefore noticeably slower than server-side services - a much stricter per-device length cap applies here than for the other video tools. Uses the compact "tiny" model tier: good for clear speech, weaker with heavy background noise or several people talking at once. Subtitles are machine-generated - review before publishing, especially proper names and numbers.
No, audio extraction and recognition run entirely locally in the browser.
SRT is the most widely supported format for video editors; VTT is the web standard for HTML5 video players.
Yes, the "Shift subtitles" tool can shift the whole file by a fixed amount.
Yes, but without speaker separation - all voices end up in the same continuous subtitle track.
Audio to text · VTT to ASS · Burn in subtitles · Video to WAV · Video to images