Auto Subtitles
VideoTranscribe speech to subtitles on your device, edit them, and burn them into the video.
Auto Subtitles listens to a video and writes the captions for you, then lets you fix whatever it misheard. The speech recognition runs on your own machine — a model is downloaded once and cached, and the video itself is never uploaded, which matters when the footage is an unreleased episode or a client's rough cut. Edit any line, restyle the text, then export SRT or VTT for a player that supports them, or burn the captions into the picture for the platforms that don't.
Why use Auto Subtitles
100% private
Your video files are processed locally and never uploaded.
Instant
No upload or download wait — it runs at the speed of your device.
Free to start
Use it free every day; go Pro for unlimited runs.
No install
Works in any modern browser. No sign-up to begin.
How to use Auto Subtitles
- 1Add the video
Drop in a clip or an audio file. The first run downloads a speech model once — after that it works offline.
- 2Generate and correct
Pick a language or let it detect one, then fix any line it misheard. Names and technical words are where it slips most.
- 3Export or burn in
Download SRT or VTT for a player that reads them, or render the captions into the picture for somewhere that doesn't.
Frequently asked questions
Is my video uploaded to transcribe it?
No. The speech recognition runs on your own device. What does get downloaded is the model itself — about 200 MB, once, then cached by your browser so later runs work with no connection at all. Your footage never leaves the machine, which is the entire reason it works this way rather than calling a cloud transcription service.
How accurate is it?
Good on clear speech, less good on names, jargon, heavy accents and crosstalk. That is why every line is editable: the tool is meant to save you typing the transcript, not to be right unsupervised. Picking the spoken language instead of leaving it on detect usually helps.
Which languages does it handle?
The model is multilingual, including Norwegian, Swedish, Danish, German, Spanish, French and others. Leave it on automatic detection or choose the language directly — naming it tends to give better results than letting it guess, especially on short clips.
What's the difference between exporting SRT and burning captions in?
An SRT or VTT file sits alongside the video, so viewers can turn captions off and the picture is untouched. Burning them in draws the text onto the frames and re-encodes, which is what social platforms that ignore subtitle files need — at the cost of a new file and captions nobody can switch off.
How long does it take?
Transcription is usually faster than the clip itself, and considerably faster on a machine whose browser supports WebGPU. Burning captions in takes roughly as long as the video runs, because it plays the whole thing through to re-encode it.