Remember what each person on your team said

Name a voice once

There is no enrollment step. Voices are separated automatically, and unnamed voices appear as placeholder people until you connect them to a real contact. Name a voice once and Audria retroactively relabels that person across your past memories, telling you how many were updated. A voice still being learned shows a "still learning this voice" state that resolves on its own.

A page for every person

Each named person has a page showing their display name, how many conversations you have shared, when you last spoke, and a playable sample of their voice. You can rename a person, remove an identification, or merge two voices that turn out to be the same teammate.

Ask, per person

The Ask tab answers questions over your own memories: what a specific teammate said about the launch, what you and a colleague discussed last week. Answers stream in with follow-up suggestions, and when Audria is unsure it asks one clarifying question instead of guessing. Conversations on video calls join the same memory through the Mac app, with no bot on the call.

What if Audria labels a voice wrong?

You can rename the person, remove the identification, or merge two voices from the person page. A voice that is still being learned cannot be force-named; it shows a learning state that resolves on its own.

Is Audria free to use?

Yes. Audria is free during early access. You can download it for iPhone and Mac today from the App Store.

Does Audria need a meeting bot or extra hardware?

No. Audria never joins your calls and needs no extra device. It runs on the iPhone and the Mac you already have.

Which tools does Audria integrate with?

Gmail, Google Calendar, Slack, Jira, Linear, Confluence, Google Contacts, Google Drive, Google Docs, Google Tasks, and Apple Reminders, with more coming.

Does anything send without my approval?

No. Every draft waits behind a confirm step and shows the exact words it came from. You review, then send.