Talk and dictation
Speak into your Mac and get text where your cursor is, plus a voice memo saved to disk.
Talk records your microphone only. Use it to dictate into any app, or to keep a quick voice memo.
Start and stop
Press ⌘⇧D from anywhere. You can also choose Talk in the launcher, or Start or Stop Recording in the menu bar, which uses the same shortcut.
- Tap the key to start, and tap it again to stop.
- Hold the key for more than a moment (350 ms) for push-to-talk. Letting go stops it.
While you talk, a small bar shows that Eavesy is listening. Nothing else on your screen moves.
Where the text goes
When you stop, Eavesy writes the words down on your Mac using macOS speech recognition. If your cursor is in a text field, Eavesy pastes the text there. Pasting needs the Accessibility permission.
This on-device text is what gets pasted, with or without an account. If on-device recognition isn't available and you're signed in, Eavesy pastes the cloud transcript instead.
When you're signed in, the audio is also uploaded and gets a cloud transcript you can find on the web. Any cleanup, such as removing filler words, happens only in that cloud copy, never in the text pasted on your Mac.
Use fn as the Talk key
You can set Talk to the fn key under Settings → Shortcuts. This needs the
Input Monitoring permission.
If pressing fn opens the emoji picker, open System Settings → Keyboard
and set "Press 🌐 key to" to Do Nothing.
What gets saved
Each Talk capture is saved as a folder named like
2026-06-15 Voice memo, with the audio and a transcript. See
your files for the full layout.
Next
These docs are open source. Spotted something wrong or missing?