§ What it does

After a meeting you need a transcript, a summary, the action items, and a follow-up email, and writing them by hand is slow. EchoScribe records the meeting and writes all four.

On newer Macs it does the work on-device with Apple Intelligence, so the audio never leaves the machine. On older Macs, or when you want speaker names, you can switch to Gemini or OpenAI with your own API key. Recording never needs a key, and the audio is saved before any analysis runs, so a failed analysis never loses a meeting.

§ Features

One-click recording with a live level meter. No account and no setup.
A transcript, summary, action items, follow-up email draft, and a mood tag for every meeting.
A local history: reopen, play back, rename, re-analyze or delete any meeting.
Sends action items to Apple Reminders in one click.
Three analysis engines: Apple on-device, Gemini, and OpenAI.
Speaker names with Gemini or OpenAI.
Signed and notarized, with no outside dependencies.

§ At a glance

Records on	macOS 14 or later
On-device analysis	macOS 26 · Apple Intelligence
Version	2026.8 · Aug 2, 2026
Price	Free
License	MIT
Account needed	No

§ Privacy

With the Apple engine, audio never leaves your Mac.

With Gemini or OpenAI, the audio and transcript go to that provider under your own API key.

Meetings are stored locally, and deleting a meeting in the app deletes its audio.

API keys are kept in the macOS Keychain.

§ Questions

Does EchoScribe need an API key?

Not to record, and not for the Apple on-device engine. Gemini and OpenAI need your own key.

Does my audio leave my Mac?

Not with the Apple engine. With Gemini or OpenAI, the audio and transcript are sent to that provider under your key.

Does it tell me who said what?

Gemini and OpenAI give speaker names. The Apple engine doesn't, but you can re-analyze a meeting with a cloud engine later.

Which Macs can run it?

Any Mac on macOS 14 or later can record. On-device analysis needs macOS 26 with Apple Intelligence turned on.