Spok3n: on-device voice dictation for Mac and iPhone
Voice dictation that runs entirely on your device — a native Mac app plus an iPhone app and keyboard. Speak, and your words appear, transcribed and AI-polished on-device. No cloud, no subscription required.
On this page
Spok3n (formerly WhisperDictate) is voice dictation that runs entirely on your device — a native Mac app, plus an iPhone app and keyboard. Speak, and your words appear, transcribed and optionally AI-polished without ever leaving your Mac or iPhone.
Under the hood it uses WhisperKit, a Swift framework that runs OpenAI's Whisper speech-to-text model directly on your device using Apple's Neural Engine. An optional AI polish step — cleaning up filler words, punctuation, and formatting — also runs on-device. It's all optimized for Apple silicon, so transcription is fast and 100% local. No internet, no cloud, no API keys.
If you want the mechanics rather than the summary, Spok3n has a technical writeup of how on-device transcription actually works — the three compiled CoreML models involved, what each one does, and how model size trades off against speed. There is also an honest comparison of on-device versus cloud dictation, including the cases where a hosted service is genuinely the better choice.
On Mac, press the Globe key → speak → your words are transcribed locally in a couple of seconds and pasted wherever your cursor is. Spok3n also keeps a searchable clipboard history of everything you've dictated.
01How It Works
Hold Mode
Hold the Globe key (bottom-left of the keyboard) → speak → release → text appears wherever your cursor is.
Tap Mode
Quick-tap the Globe key → speak as long as you want → tap again → text appears. Great for longer dictation.
Spok3n Keyboard
In any app, switch to the Spok3n keyboard and tap Dictate. Spok3n records and transcribes on-device, then drops the polished text right back into your text field. Your keystrokes never leave your phone.
On Mac, a small colored bar appears at the bottom of your screen to show status:
02Install on Mac (~2 minutes)
- Download Spok3n from spok3n.com, open the DMG, and drag the app to Applications
- Open Spok3n (it’s signed and notarized by Apple, so it opens normally)
- The speech model downloads automatically (~142 MB, takes about 20 seconds)
- Grant Accessibility permission when prompted —
System Settings → Privacy & Security → Accessibility, and toggle Spok3n on - Grant Microphone access when prompted
- Done! Look for the waveform icon in your menu bar
03Tips
- Click the waveform icon in the menu bar to see status and settings
- Enable "Launch at Login" from the menu bar dropdown so it's always ready
- Pick an accent color in Settings — it recolors the whole app to match your style
- If you see a instead of the waveform, Accessibility permission needs to be granted — click the icon for instructions
- Works best on Apple Silicon Macs (M1/M2/M3/M4) — Intel works but transcription is slower
Privacy First
All transcription and AI polish happen locally on your device using WhisperKit (Apple CoreML + Neural Engine). No audio or text is ever sent to any server, and the Spok3n keyboard never transmits your keystrokes. Read the full privacy policy.
Spok3n is one of Apex’s own products. For small businesses, Apex builds websites and business automation with the same attention to privacy and security.
