Local video in. Synced subtitles out. Nothing leaves your browser.
⏺ 0 network uploadsMP4, WebM, MOV — decoded and transcribed entirely on this device (WASM + Whisper)
First run downloads model weights from the Hugging Face Hub and caches them in your browser (IndexedDB). Every run after that works fully offline. Longer videos and larger models take proportionally longer — this all runs on your CPU/GPU, not a server.