Caption generator
Generate subtitles for any video or audio with AI — right in your browser. Export SRT, VTT, or plain text. Nothing is uploaded.
Pick a video or audio file. The AI model downloads once (~40–120 MB) and is cached for future visits.
How it works
Add your file
Video or audio — transcription runs in 30-second chunks, so even long files work.
Generate captions
Whisper AI transcribes with timestamps. Fix any word in the editor before exporting.
Export
Download SRT (editors, YouTube), VTT (web players), or plain text.
Frequently asked questions
Is my file uploaded?
No. Only the AI model files download once from a public CDN. Your media never leaves your device.
How accurate is it?
Whisper is among the most accurate open models. Clear speech needs few fixes; noisy audio benefits from the Accurate preset.
Which languages?
About 99, including English, Hindi, Spanish, French, German, Portuguese, Arabic, Chinese, Japanese and Korean.