Video to text
Choose a video and press Transcribe: the sound is taken from the video and an AI speech-recognition model writes down what is said, all inside your browser. Copy the transcript from the page or download it as a text file. Need captions with timings instead? Pick SRT or VTT, or use the Subtitle generator.
It reads MP4, MOV, WebM and MKV files, including screen recordings, phone videos and exported meetings. The language is detected automatically, or you can choose it.
Good to know
Clear speech close to the microphone gives the best transcript. Background music, noise and crosstalk lower the accuracy, so read the text before you share it. The picture isn't used: only what is said ends up in the transcript.
Private by design
Your video stays on your device: nothing is uploaded, so long recordings of meetings or classes don't have to travel anywhere. The first time, the AI model (about 136 MB) is downloaded and kept by your browser. Videos up to 90 minutes long.
How this works
The video's sound track is decoded and converted to 16,000 samples per second, then split into 30-second windows. The Whisper base model (OpenAI, MIT licence) reads a spectrogram of each window and writes the words with timestamps; windows restart where a sentence was cut, so no words go missing between them.
Frequently asked
Is my recording uploaded?
No. Speech recognition runs in your browser on this device; the file never leaves it.
Which languages work?
Around 100, including English, Portuguese, Spanish, French and German. It detects the language by itself, or you can choose it.
How long can the recording be?
Up to 90 minutes per file. A laptop transcribes roughly six minutes of audio per minute; the first time also downloads the AI model (about 136 MB).
How accurate is it?
Very good for clear speech by one person at a time. Background noise, music, strong accents and people talking over each other lower the accuracy, so read the text before you use it.
Does it read text shown in the video?
No, only speech. For text in a picture, take a screenshot and use Image to text.
Can it go faster?
Yes, with a free account: choose Fast, in the cloud. Whisper large-v3-turbo transcribes an hour of audio in a few minutes for about 0.5 credit a minute (free accounts get 50 credits a month). The audio is sent in two-minute pieces and not stored. On this device stays free and private.
Sources
- OpenAI Whisper — robust speech recognition via large-scale weak supervision (MIT)
- Radford et al. (2022) — Robust Speech Recognition via Large-Scale Weak Supervision
- ONNX Runtime Web — run ONNX models in the browser (WebGPU and WebAssembly)
- Mediabunny documentation — Converting media files
- W3C — WebVTT: The Web Video Text Tracks Format
- Wikipedia — SubRip
Reviewed Oct 19, 2026 · test vectors fixtures/tools/video/transcribe-video/