Online Subtitle Generator — Local AI Speech-to-Text (Whisper)

Generate SRT and VTT subtitles for audio or video with local Whisper speech recognition. The AI model runs in your browser — files never upload.

Try it Now →

Why Choose Formavide?

  • Works in Chrome, Edge and desktop Safari; mobile handles clips up to 3 minutes.
  • Privacy first: the AI runs on your device — your audio and video are never uploaded.
  • Real models, not toys: production-grade neural networks running right in your browser.
  • Every AI tool comes with a free daily quota — no subscription; a one-time purchase lifts the cap.

How to Use

  1. Upload an audio or video file (MP3, WAV, M4A, MP4, WebM and more).
  2. Pick a model — Whisper Base is faster (~75MB first download), Whisper Small is more accurate (~250MB).
  3. Click start. Recognition runs locally in your browser (model downloads on first use).
  4. Preview and edit the subtitles, then export as SRT or VTT.

FAQs

How fast is the transcription?

It depends on your device. Browsers with WebGPU get GPU acceleration automatically; otherwise it falls back to CPU, which is slower.

Is my audio private?

Yes. The Whisper model runs in your browser — audio never leaves your device.

Why is there a model download on first use?

The Whisper model is fetched from a public CDN (Base: ~75MB, faster to download; Small: ~250MB, more accurate) and cached by your browser afterwards.

Which languages are supported?

Language auto-detection is on by default, with manual Chinese/English override, based on the Whisper model.

Can I burn subtitles into the video?

The current version exports SRT/VTT subtitle files; burned-in captions are not supported yet.

More Free Online Tools

View all tools →
Local-first: your media never leaves this device.