Speak. It types.
Lightning-fast voice-to-text for macOS. Press a hotkey, speak, and your words appear wherever you're typing.
Free & open source · Bring your own free Groq key
Let AI set it up for you
Paste this into Claude Code (or any AI assistant that can run commands on your Mac) and it'll download, install, and walk you through setup.
Download and set up the Dictate app on my Mac: 1. Download https://github.com/oliwoodman/Dictation/releases/latest/download/Dictate.zip 2. Unzip it and move Dictate.app into /Applications 3. Open it 4. Then walk me through the first-run wizard: I'll grant Microphone and Accessibility permissions when it opens System Settings, and I'll paste a free Groq API key from https://console.groq.com/keys It's signed and notarized by Apple. To use it after setup I hold Right Option + Space, speak, and let go.
Set up in 3 steps
Download & install
Grab the latest Dictate.zip from GitHub, unzip it, and drag Dictate.app into your Applications folder. It's signed & notarized by Apple, so it opens cleanly.
Get a free Groq API key
Dictate is bring-your-own-key. Create a free key at console.groq.com/keys and paste it into the setup wizard. Your audio goes straight to Groq — nowhere else.
Press, speak, done
Hold Right Option + Space, talk, and let go. Your words paste into whatever app you're in.
Instant transcription
Powered by Groq's Whisper API. Your speech appears in under a second.
Works in every app
Slack, Chrome, VS Code, Notes — anywhere you can type, Dictate pastes your words.
Private by design
Bring your own free Groq key. Audio goes straight from your Mac to Groq — no account, no server, no telemetry.
Free & open source
MIT licensed. No subscription, no catch. Groq's free tier handles everyday dictation.
Heads up: Dictate isn't actively maintained. I built it for myself and use it daily — it works, but it has rough edges and I'm not managing bugs or feature requests. It's shared as-is, for free. The code is open source on GitHub — fork it and make it your own.
macOS 12.0+ · Native app · Under 5MB
Setup help & troubleshooting in the GitHub README