Video to Text
Transcribe a video's speech to text and subtitles (SRT/VTT) with Whisper — in your browser, nothing uploaded.
Select an audio or video file
MP3, WAV, M4A, MP4, MOV, WebM — transcribed in your browser with Whisper
Click to choose, or drop a file here
About transcribing video in the browser
This turns the speech in a video into text — and into SRT or VTT subtitle files — without uploading anything. It runs OpenAI's Whisper model locally in your browser through WebAssembly, so the audio is decoded from your video, resampled, and fed to the model entirely on your machine.
There's a one-time download the first time you pick a model size: roughly 40 MB for Tiny and 80 MB for Base (both quantised). After that the browser caches the weights, so it starts instantly and even works offline. Base is the better default for clear speech; Tiny trades accuracy for speed and is handy for a quick rough pass or a slow device. Whisper adds its own punctuation and capitalisation, auto-detects the language (or you can set it), and can translate non-English speech straight to English.
It's not a replacement for a human transcriptionist: expect mistakes on strong accents, overlapping speakers, background music and low-quality audio, and note that it doesn't label who's speaking. For subtitles, the SRT/VTT timings come from Whisper's segment timestamps and are usually close enough to use directly, sometimes with light nudging.
Supported files
Accepts MP4, MOV, WebM and other video your browser can play (it needs an audio track). Exports .txt, .srt and .vtt.
Is my file uploaded?
No. Whisper runs entirely in your browser via WebAssembly — the video and its audio never leave your device.
Common problems
- "Couldn't read audio from this file"
- The video may have no audio track, or your browser can't decode its audio codec. Extract the audio first (Video to MP3) and transcribe that.
- It's very slow
- WASM transcription runs at roughly 1–5× real time. Switch to the Tiny model, or trim the video to the part you need first.
- The text has errors
- Whisper isn't perfect — proofread. Setting the exact language instead of Auto-detect, and using the Base model, both help.
How to Video to Text
- 1
Add your video
Drop in an MP4, MOV or WebM with an audio track — it's read in your browser, never uploaded.
- 2
Pick a model and language
Base is more accurate, Tiny is faster. Leave the language on Auto-detect or set it. The model downloads once (~40–80 MB) then it's cached.
- 3
Transcribe and export
Click Transcribe; when it's done, copy or download the transcript as plain text, SRT or VTT subtitles.
Related PDF tools
Video to Text — frequently asked questions
Is my video uploaded?+
No. The transcription model (OpenAI's Whisper) runs entirely in your browser via WebAssembly. The video never leaves your device.
Why is there a one-time download?+
The Whisper model weights (~40 MB for Tiny, ~80 MB for Base, quantised) download the first time you use that size, then the browser caches them — after that it works offline and starts instantly.
How accurate is it?+
Base is solid for clear speech in a supported language; Tiny is noticeably rougher but much faster. Both struggle with heavy accents, crosstalk, music beds and poor audio — proofread the output.
What can I export?+
Plain text, or timed subtitles as SRT or VTT. The SRT/VTT timings come from the model's segment timestamps.
Do you upload my files to a server?+
No — every tool here processes your files entirely on your device, and nothing is uploaded to a server. The only exception is the AI tools, which send the extracted text (not the file) to Claude to do their work.
Is there a file size limit?+
There's no artificial limit we impose, but since processing happens in your browser, very large files (500+ MB or 1,000+ pages) are limited by your device's memory rather than by us.
Advertisement
Advertisement