Getting started
Cadenza turns speech into text where your cursor is. Hold a shortcut, speak, release.
What you need
Section titled “What you need”- A Mac (Apple silicon or Intel) running macOS 14 or later.
- About 200 MB of free space for the recommended local model (the app downloads it when you ask).
- A microphone.
1. Install
Section titled “1. Install”Download the disk image from GitHub Releases, open it and drag Cadenza to Applications. The app is not notarized, so macOS blocks the first launch: open System Settings → Privacy & Security and choose Open Anyway. You can also build it from source in a few minutes.
2. First run
Section titled “2. First run”A short wizard opens the first time. It explains the privacy promise, guides you through the permissions and lets you choose how speech is recognized. The sensible choice is a local model: recognition happens on your Mac and nothing is uploaded. SenseVoice is recommended; the app downloads it for you.
3. Allow the permissions
Section titled “3. Allow the permissions”| Permission | Why |
|---|---|
| Microphone | Capture your speech while you record |
| Accessibility | Find the text field and type the result |
| Input Monitoring | Detect the global shortcut |
| Speech Recognition | Only if you choose Apple’s built-in recognition |
Settings → Privacy shows the state of each one. If the shortcut page says a permission is “on but not working”, see Shortcuts and permissions.
4. Dictate
Section titled “4. Dictate”- Click into any text field.
- Hold your shortcut (Left Option by default), speak, then release.
- The text is typed at the cursor. If it cannot be typed, the app keeps it so you can copy it.
Press Esc to cancel a recording. You can change the shortcut, or use “tap to start, tap to stop”, on the Shortcut page.