Turn a video or audio clip into subtitles and a transcript without uploading anything. Drop a file and the tool decodes its audio, runs OpenAI’s Whisper speech-recognition model right inside your browser, and gives you timed captions you can download as .srt or .vtt for any video editor or player, plus a plain-text transcript.
Because the model runs on your device, your recording never leaves your computer — a real difference from online transcription services that upload your file to a server. It works offline once the model has loaded, and handles many languages, with an option to translate speech to English.
No. The audio is decoded and transcribed entirely in your browser using an on-device AI model. Your file never leaves your device.
SubRip (.srt) and WebVTT (.vtt), which every major video editor and player supports, plus a plain-text (.txt) transcript.
Whisper is multilingual and auto-detects the spoken language. You can also translate non-English speech to English subtitles with one toggle.
The AI model downloads once (from this site, or a public source if it is not self-hosted) and is cached by your browser; after that it loads instantly and works offline. Either way, your audio is only ever processed locally — never uploaded.
It uses a compact Whisper model that runs fast on-device; clean audio transcribes well, and you can edit the text before exporting. Cloud models can be more accurate but require uploading your file.