Privacy

Avette is built so your speech stays yours.

Audio never leaves your Mac

Dictation is transcribed entirely on-device. Avette does not record, upload, or transmit your audio or transcripts to bearcove or anyone else.

What does touch the network

• Model download: on first run, Avette downloads its speech model from our model server (models.avette.eu). As with any download, that server sees your IP address.
• License: when you activate, and from time to time after, Avette sends your license key, a random identifier for your Mac and your Mac's name to our licensing service (store.avette.eu) to confirm your purchase.
• Updates: once a day Avette checks updates.avette.eu for a new version. Nothing about you or your Mac is sent.
• iPhone link: only once you pair an iPhone; its dictation reaches your Mac encrypted, directly or through a relay that cannot read it.

That's it: no analytics, no telemetry, no account.

Read the full Privacy Policy for details on purchases, licensing, your rights, and data retention.