Verbatim
Model and files stay in this browser
Free · Private · No limits

Transcribe audio and video, privately.

Whisper runs on your device, not a server. Record in this tab or drop a file, and watch it become clean, timestamped text. Your words never leave your machine.

Tap to record

or drop a file anywhere, or

MP3 · WAV · M4A · MP4 · MOV · and more · no length limit

The whole point

Your words never leave your device.

Nothing is uploaded

There is no server to send to. Audio, video, and the finished transcript stay in this tab, full stop.

The model runs locally

Whisper downloads once and runs on your own hardware. No account, no API key, no usage being watched.

Works offline

After the first load you can pull the plug. Transcribe a sensitive call on a plane, in a clinic, anywhere.

How it works

Three steps. No setup, no sign in.

01

Record or choose

Hit record to capture from your mic, or drop in an audio or video file of any length.

02

It transcribes on your device

Whisper runs locally and turns speech into clean, timestamped text. Nothing is sent anywhere.

03

Edit and export

Fix a word, switch to timestamps, then copy or save as .txt, .srt, or .vtt.

Why Verbatim

Built to be quiet, fast, and yours.

Free, no length limits

Transcribe a two minute note or a three hour interview. There is no meter and no paywall.

Fully private

Everything runs in your browser. Nothing is uploaded, logged, or trained on.

Many languages

Transcribe in dozens of languages with auto-detect, or translate speech into English.

Timestamps and subtitles

Every segment is timed. Export caption-ready .srt and .vtt in a click.

Works offline

After the first load Verbatim keeps working with no connection at all.

No account

Open the page and start. No email, no password, nothing to forget.

Who it is for

For anyone turning speech into text.

Interviews

Keep sources confidential. Nothing leaves your laptop.

Lectures

Capture every word, then study from clean text.

Meetings

Minutes without a bot in the call or a cloud recording.

Voice notes

Talk it out, get text you can search and paste.

Captions

Make video accessible with subtitle files in seconds.

Honest answers

Questions, answered plainly.

Yes. Verbatim has no backend to receive your files. The model and your audio live in your browser tab. You can confirm it yourself: open your network tools, transcribe a file, and watch that nothing is sent.
There is no hard limit. Length is bounded only by your own device memory and patience. Very long files take longer because the work happens on your hardware, not a data center.
Plain text (.txt), SubRip subtitles (.srt), and WebVTT (.vtt). You can also copy the transcript straight to your clipboard, with or without timestamps.
It uses OpenAI's Whisper models, which are strong across clear speech and many languages. Choose a larger model for better accuracy, or a smaller one for speed. The transcript is fully editable, so you stay in control of the final text.
It is free, and there is no catch. The honest trade-offs: the model downloads once on first use (a one time wait), and because everything runs locally, very long files are slower than a paid cloud service. For most people, privacy and no limits are worth it.