How we handle your audio and data. Plain and honest.
Your uploaded audio is sent to our speech-to-text provider to make a transcript, then the audio file is deleted from our server right after processing. We don't keep your audio files.
Your transcript and AI summary are stored in your account so you can read them later. Only you can see your own notes.
We use third-party speech-to-text and language-model providers to transcribe and summarize. Your content is processed to provide the service and is not used to train their models (per provider settings). We don't sell your data.
We use your email only to send your sign-in link and account notices.
If you connect an email inbox, we read it read-only over IMAP to show your unread mail — we never mark mail as read, and never send a reply without your explicit confirmation. Your app password is encrypted at rest and only decrypted to make the connection. The text of those emails is sent to our language-model provider to summarize, the same way audio transcripts are.
If you connect a calendar, we write your deadlines to it. With Apple or Yahoo we use an app password over CalDAV; with Google we use OAuth. App passwords and OAuth tokens are encrypted at rest and only used to add your events. You can also subscribe to your deadlines with a private subscription link (.ics) — treat it like a password. You can turn it off (revoke) at any time, and disconnecting a provider disables its stored credentials.
You can ask us to delete your account and all stored notes at any time.