Listen
Remember what was said.
No bot joins your call. Your recordings stay on your Mac.
This looks like an Intel Mac, and Listen will not run on it.
Listen transcribes on your own machine, and that needs the Neural Engine Apple introduced with the M1 in late 2020. There is no Intel version, and there is no version that would work.
Open the Apple menu and choose About This Mac. If it says Chip rather than Processor, this guess is wrong and you can download Listen as normal.
A demo meeting. The people and voices are synthetic.
Stay in the meeting
Listen records both sides of the call. When you need something later, play it back and read along.
Keep your own notes
Write down what matters while you listen. When you want a summary, your assistant can make one from the meetings you choose.
A demo of personal notes and a summary across two meetings.
Dictate into anything
Press fn + left shift wherever you are typing, say what you want written, and press it again. The words are typed in where the cursor is: an email, Slack, a search bar, a code comment.
They go on the clipboard every time as well, so ⌘V
still gets you the words when the app in front will not accept a keystroke
from Listen. Escape cancels, and so does the trash button on the floating
pill.
It is the same speech model that transcribes your meetings, running on the same Mac, so there is nothing extra to download and nothing you dictate is uploaded.
Dictating needs Accessibility, which is what lets Listen see the shortcut and type for you. Recording meetings never uses it, so if you only want the recorder you can ignore this and Listen will not ask.
Your meetings stay on your Mac
There is no meeting bot and no account. Your audio, transcripts and voiceprints never leave your Mac.
Listen downloads its speech model once and checks for updates. It does not upload your meetings or collect telemetry.
Catch the start
Press Record when the meeting begins. Decide whether to keep it when you are done.
Coming from Speak
Dictation used to be Speak, a separate app. It is part of Listen from 0.12.0, and Speak 1.6.0 is its last release.
The speech model carries over on its own: both apps always used the same download, so there is nothing to fetch again. The configuration does not. Set the shortcut again in Settings, Dictation, and bring your Speak dictionary across in one press from Settings, Dictionary.
Your copy of Speak keeps working. It will not be updated again.
Install with Homebrew
brew trust --cask mugoosse/tap/listen
brew install --cask mugoosse/tap/listenApple Silicon, macOS 14 or later. System audio capture needs macOS 14.2; below that Listen records your microphone only.
The speech model is about 2.5 GB and downloads on first run. If you use Speak already, Listen uses the copy on your Mac.