About CaptionFlow

CaptionFlow is an automatic subtitle generator powered by OpenAI's Whisper model — running entirely in your browser. Your video file never leaves your device.

How it works

1. You drop a video or audio file. The browser reads it locally.
2. FFmpeg.wasm extracts a 16 kHz mono WAV track.
3. Whisper (running via Transformers.js + WebAssembly) transcribes the audio.
4. You download the subtitles as SRT, VTT, TXT, or JSON.

Models

We support multiple Whisper sizes — tiny (39M params, fastest), base (74M, balanced), and small (244M, most accurate). Bigger models are slower but produce better transcriptions.

Languages

Whisper supports 99 languages. CaptionFlow defaults to English-only (faster) but you can switch to multilingual mode for any other language.

Free forever

No upload fee, no per-minute charge, no watermark, no account. The cost to us is zero (your device does the work) so the cost to you is zero.