Your voice stays on your Mac.

OpenDictation is built around a simple principle: your data is yours. Everything runs locally on your laptop — no cloud, no servers, no third parties.

100% on-device

All speech recognition happens directly on your Mac. Your audio never leaves your machine — not even for a millisecond. There are no network requests, no API calls, no telemetry. Just your voice and your laptop.

Powered by Voxtral

OpenDictation uses Mistral's Voxtral speech model, running entirely on your hardware. More model options are coming soon.

No data collection

We don't collect your audio, your transcriptions, or any usage data. There are no analytics, no tracking pixels, no cookies. We have no servers to send your data to — even if we wanted to, we couldn't.

Your notes are yours

Every word you dictate stays on your computer. Your transcriptions are injected directly into whatever app you're using — they're never stored, logged, or cached by OpenDictation. When you close the app, there's no trace.

No account required

No sign-up, no email, no login. Download it, open it, use it. That's it.

Open source

Don't take our word for it — read the code. OpenDictation is fully open source. You can verify every line, audit every dependency, and build it yourself.

Last updated: March 2026. If you have questions, open an issue on GitHub.