AI Whisper · runs in your browser

Audio to Text — Free Transcription

Upload an MP3, WAV, or M4A and get an accurate transcript — powered by an AI model that runs entirely on your device.

Drop an audio or video file, or click to browse

MP3, WAV, M4A, OGG, WEBM, MP4… processed locally in your browser

Model

0 words

Private: the AI model runs entirely in your browser. Your file is never uploaded to any server.

Voice to Text Dictation Voice Notes Audio to Text Subtitle / SRT Meeting Transcription

Turn recordings into text without sending a single byte to a server. This tool loads OpenAI's Whisper speech model into your browser (via WebAssembly) and transcribes your audio file locally. Works in any modern browser. Export plain text or timestamped subtitles.

Any audio file

MP3, WAV, M4A, OGG, WEBM and more.

On-device AI

Whisper runs locally — your file is never uploaded.

Timestamps

Export a .srt subtitle file with time codes.

No limits

Free, no sign-up, no length cap (longer files take longer).

Frequently asked questions

Everything you need to know about this free tool.

Is this voice-to-text tool really free?▾

Yes — completely free, unlimited, with no sign-up, no watermark, and no time limit. There is nothing to install and no credit card required.

Does my audio get uploaded to a server?▾

No. Live dictation uses your browser's built-in speech recognition, and the audio-file transcriber runs an AI model (Whisper) directly inside your browser via WebAssembly. Your voice and recordings never leave your device — there is no upload.

Which browsers are supported?▾

Live microphone dictation works best in Google Chrome and Microsoft Edge (and Chrome on Android). The audio-file transcriber runs in any modern browser, including Firefox and Safari, because it does not depend on the browser's speech engine.

How accurate is the transcription?▾

Accuracy is very good for clear speech in a quiet room with a decent microphone. Speaking at a natural pace, avoiding heavy background noise, and using a headset mic all improve results. You can always edit the transcript directly in the box.

Can I add punctuation by voice?▾

Yes. Say commands like “period”, “comma”, “question mark”, “new line”, or “new paragraph” and the tool inserts the right character automatically.

Can I export the text?▾

Yes — copy to clipboard with one click, or download as a .txt file. The audio-to-text and subtitle tools can also export timestamped .srt subtitles.

How it works & your privacy

The technology

Live dictation uses your browser's built-in Web Speech API.
Audio-file transcription runs OpenAI's Whisper model in-browser via WebAssembly (transformers.js).
No backend, no API keys, no usage limits.

Your privacy

Audio and recordings never leave your device.
Transcripts auto-save only to this browser's local storage.
No sign-up, no tracking of your content.

Free to use. Accuracy varies with audio quality, accent, and background noise. Always review transcripts before relying on them.

Loading Legacy