Spoken is voice-to-text that runs entirely on your Mac. No audio leaves your device. No subscriptions. No compromises.
Requires Apple Silicon (M1 or later) · macOS 14.0+ · Intel Macs are not supported
Features
Built for speed, privacy, and reliability. Spoken stays out of your way and just works.
Spoken uses WhisperKit locally on Apple Neural Engine. No audio leaves your Mac, ever. Works fully offline.
Download a small LLM (Phi-4 Mini, Gemma 3, etc.) to polish grammar and formatting entirely on-device. No API key needed.
Add custom terms and aliases so Whisper always gets your names, brands, and technical terms right. Tag UI with newline-separated aliases.
When you re-dictate a phrase, Spoken offers to save the correction to your dictionary automatically.
Searchable history of everything you've dictated and copied. Cross-device sync via your account.
Auto-detect context — email, code, chat, or notes — and format your text accordingly.
Say "new paragraph", "period", or "scratch that" to control formatting without touching the keyboard.
Manage your personal dictionary from the terminal with the spoken CLI.
What's New
The latest release — always up to date.
Download a small local LLM (Phi-4 Mini, Gemma 3, and more) to clean up grammar, punctuation, and formatting — all on-device. No API key, no cloud, no subscription.
Redesigned tag/chip UI for managing custom terms. Add multiple aliases per term (newline-separated). Import from URL or share dictionaries with teammates.
Re-dictate a phrase and Spoken detects the correction, offering to save it to your dictionary so the same mistake never happens twice.
Manage your personal dictionary from the terminal. Add terms, list aliases, and batch-import — perfect for developers who live in the command line.
Most dictation apps send your audio to remote servers for processing. Spoken takes a fundamentally different approach. All transcription and text polishing happens locally on your Mac using Apple's Neural Engine. No audio or text data is ever transmitted, stored, or processed externally.
Compare
The only dictation app that combines local processing with a one-time purchase.
| Spoken | Wispr Flow | Aqua Voice | SuperWhisper | |
|---|---|---|---|---|
| Pricing | Pay once | $15/mo | $8/mo | $10/mo |
| Local processing | Yes | Cloud | Cloud | Yes |
| Offline support | Yes | No | No | Yes |
| Session limits | None | Varies | Varies | None |
| Cross-device sync | Yes | No | No | No |
| Personal dictionary | Yes | No | No | No |
| On-device text polishing | Yes | Cloud AI | Cloud AI | No |
| Audio data privacy | Never sent | Sent to cloud | Sent to cloud | Local |
Account
Step 1: Download Spoken and open the DMG.
Step 2: Drag Spoken to your Applications folder.
Step 3: First launch: Double-click to open. If macOS blocks it, go to System Settings → Privacy & Security, scroll down, and click Open Anyway next to the Spoken message. This is only needed once.
Step 4: Create an account below to unlock full features and clipboard sync.
Requires Apple Silicon (M1, M2, M3, or M4). Intel Macs are not supported.
Alternative: run in Terminal: xattr -cr /Applications/Spoken.app
Or: Right-click the app → Open → click Open again in the dialog.