Permissions

dictate needs three macOS permissions to record audio and insert text into other apps.

Accessibility

Accessibility lets dictate observe the focused app, read selected text when available, and send paste keystrokes.

1. Open System Settings.
2. Go to Privacy & Security → Accessibility.
3. Click + or enable the existing entry for Terminal, your packaged dictate.app, or the Python runner you use during development.
4. Restart dictate after granting access.

If selection-as-context does not work, confirm the focused app exposes selected text through macOS Accessibility APIs.

Microphone

Microphone permission lets the recorder capture your voice.

1. Open System Settings.
2. Go to Privacy & Security → Microphone.
3. Enable Terminal, your packaged dictate.app, or the Python runner.
4. Relaunch dictate and run the onboarding mic test or dictate doctor.

When packaged as an app, dictate includes NSMicrophoneUsageDescription so macOS can show the permission prompt.

Input Monitoring

Input Monitoring is required for the global hotkey and synthetic keyboard events used for reliable insertion.

1. Open System Settings.
2. Go to Privacy & Security → Input Monitoring.
3. Enable Terminal, your packaged dictate.app, or the Python runner.
4. Quit and reopen dictate.

Without Input Monitoring, the hotkey may not fire or paste insertion may fail in focused apps.