Video to Text
Transcribe any video or audio file to text with Whisper AI — right in your browser, so the file is never uploaded. Get readable paragraphs, timestamps, or SRT/VTT subtitles, in 90+ languages or translated to English.
Input
Runs entirely in your browser — the file is never uploaded. The model downloads once, then is cached for next time.
Transcript
Upload a video or audio file and press Transcribe. The transcript will appear here.
Also available with
Guides
What this tool does
Upload a video (MP4, MOV, WebM…) or an audio file and get a written transcript of everything that's said in it. The transcription runs on your own device with OpenAI's open-source Whisper speech-recognition model — the file is never uploaded, so it works for private recordings, client calls and unreleased footage.
How to use it
- Drop in a video or audio file.
- Pick Speed vs. accuracy — Fast for a quick draft, Balanced for most recordings, Accurate for difficult audio, accents or technical vocabulary.
- Optionally set the spoken language (otherwise it's detected), or choose Translate to English to get an English transcript of foreign-language speech.
- Press Transcribe. A progress bar shows the model download and then each 30-second segment as it's processed.
- Pick an output format — readable paragraphs, plain text, timestamped lines, an SRT or WebVTT subtitle file, or JSON — and copy or download it. Switching formats is instant; nothing is re-transcribed.
How long does it take?
The first run downloads the model (about 40 MB for Fast, 80 MB for Balanced, 250 MB for Accurate); after that it's cached by your browser and starts immediately. Transcription speed depends on your computer: on a typical laptop the Balanced model handles a 10-minute video in a few minutes. Fast is roughly twice as quick; Accurate is several times slower.
Frequently asked questions
Is my video uploaded anywhere? No. The audio is decoded and transcribed inside your browser tab. The only thing downloaded is the speech model itself (from Hugging Face's CDN).
Which file types work? Anything your browser can play: MP4, MOV, WebM and M4A/MP3/WAV work in every major browser; MKV and AVI depend on the codecs inside. If a file won't open, convert it with Extract Audio from Video first — the MP3 it produces transcribes fine.
How long can the file be? Up to 3 hours. Very long files need plenty of memory; if your browser runs out, try the Fast model or split the recording first with Video Cutter.
Can I make subtitles with it? Yes — choose SRT subtitle file or WebVTT subtitle file and download. Timings come from Whisper's own segment timestamps.
I have a YouTube link, not a file. Use YouTube to Text, which reads the video's existing captions without downloading anything.
Can it transcribe my microphone live? That's Speech to Text — live dictation through your browser's speech recognition.