How it works

FastSub AI is a static website. When you open a tool, your browser downloads the app and runs everything locally.

Your files stay on your device

Files you drop are read with the browser's File API and processed in memory. There is no upload endpoint at all.

AI models run in the browser

• Speech recognition: OpenAI Whisper (base, small, large-v3 turbo) via ONNX Runtime Web.
• Translation: Chrome's built-in translator, or Opus-MT models.
• OCR for image subtitles: Tesseract.
• Voices: Meta MMS text-to-speech.

Models download once and stay in your browser cache. With WebGPU they run on your graphics card; otherwise on the CPU with WebAssembly.

Video processing

Audio extraction, subtitle burn-in and dubbing use WebCodecs through Mediabunny, the same hardware encoders your browser uses for video calls.