Getting started

Setup takes about two minutes. You'll need internet once, to download a speech model.

1. Install and open

Download Sayburst, drag it to Applications, and open it. A waveform icon appears in the menu bar, and the setup window opens.

2. Allow two permissions

Microphone lets Sayburst hear you while you hold the key. Accessibility lets it notice the key and paste into other apps. Setup walks you through both; if something goes wrong, see Permissions.

3. Choose your key

The default is fn (the 🌐 key on newer keyboards). If macOS is set to show emoji or change input source when you press it, setup shows how to turn that off — or choose right ⌥ instead. External keyboards often have no fn key; use right ⌥ or a custom shortcut.

4. Download a model

Sayburst recommends Parakeet v3 (about 480 MB) for English and 24 European languages, or Whisper Large v3 Turbo for other languages. The first dictation after a download takes about 20 seconds while macOS prepares the model; after that it's fast. Compare models.

5. Dictate

1. Click into any text field — an email, a chat box, a document.
2. Hold fn and speak. The overlay at the bottom of the screen shows that Sayburst is listening.
3. Let go. Your text appears where the cursor was.

A quick tap of the key starts hands-free dictation; tap again to finish. Esc cancels.

Spoken commands

In the Clean, Message, Email and Notes modes, say “new line”, “new paragraph” or “bullet point”.

If text doesn't appear

Sayburst never throws text away. If it couldn't type — because you switched apps, a password field was focused, or Accessibility is off — the overlay says so and offers Copy and Paste here. You can also click the menu bar icon to see your last dictation, press ⌃⌥⌘V to paste it, or open History. More fixes in Troubleshooting.

Next steps

• Add names and jargon in Settings › Dictionary.
• Set modes per app in Settings › Writing.
• Add shortcuts for rewriting selected text, asking questions and capturing notes in Settings › Shortcuts.