Installation Guide
macOS
Scribe requires macOS 13 (Ventura) or later. It runs on both Apple Silicon and Intel Macs.
- 1. Download the .dmg from the download section or the link above.
- 2. Drag Scribe to Applications. Open it from your Applications folder.
- 3. Grant Microphone permission. macOS will prompt you on first launch. Scribe needs this to hear your voice.
- 4. Grant Accessibility permission. Required for typing transcribed text into other applications. Go to System Settings → Privacy & Security → Accessibility.
Quick Start
Once permissions are granted:
The first transcription may take a few seconds while the Whisper model loads into memory. Subsequent transcriptions are near-instant.
Troubleshooting
Permissions not working after update
macOS may revoke permissions when the app is updated or re-signed. Go to System Settings → Privacy & Security and re-enable Scribe for Accessibility.
Slow first transcription
The Whisper model takes 20-30 seconds to load into GPU memory on first use. After that, transcriptions are near-instant. The model stays loaded while Scribe is running.
Changing Model
Scribe supports three Whisper model sizes. Larger models are more accurate but use more memory and take longer to load.
To switch models, open Settings (gear icon in the app) and tap a model card. If the model isn't downloaded yet, Scribe will download it automatically before loading.
Changing Hotkey
The default push-to-talk hotkey is Fn on Mac.
To change it:
- 1. Open Settings in the app.
- 2. Click the current hotkey button under Hotkey Selection.
- 3. Press your desired key or key combination. Scribe will capture and save it immediately.
You can use any single key or modifier combination (e.g. Ctrl+Shift+Space). Press Esc to cancel.
Related Guides
Need workflow-specific setup help? These focused guides show how developers use Scribe in coding environments.