Your Voice,
Your Mac,
Your Privacy.
100% offline voice-to-text for macOS. Hold a hotkey, speak, and text appears at your cursor. Open source, private by design, and completely free.
VocaMac does it all locally!
Three Steps. That's It.
No account creation. No cloud configuration. A guided setup wizard gets you started in minutes.
Hold Right Option
Press and hold your hotkey. The menu bar icon turns red. You're recording.
Speak
Say whatever you want. Your audio is processed entirely on your Mac.
Release
Text appears at your cursor. In any app. Browser, Slack, VS Code, anywhere.
Everything You Need,
Nothing You Don't
Built for developers and power users who value privacy and speed.
System-Wide Text Injection
Transcribed text appears wherever your cursor is. Browsers, Slack, VS Code, spreadsheets, terminals. Everywhere.
Push-to-Talk
Hold a hotkey to record. Release to transcribe. Simple, predictable, no surprises.
Double-Tap Toggle
Prefer toggle mode? Double-tap to start, double-tap to stop. Your choice.
Smart Model Selection
Auto-detects your hardware (Apple Silicon, Intel, RAM) and recommends the optimal Whisper model.
Apple Silicon Native
CoreML + Metal + Neural Engine acceleration via WhisperKit. Blazing fast on M1/M2/M3/M4.
Fully Configurable
Choose hotkeys, models, languages, silence detection. Settings window with model management.
99+ Languages
Auto-detect or specify. English, Spanish, French, German, Chinese, Japanese, and many more.
Visual Feedback
Menu bar icon changes color. Audio level indicator shows input. Cursor indicator near your caret while recording.
Translation
Speak in one language, get text in another. Built-in speech translation powered by Whisper.
Silence Detection
Auto-stops recording after you stop speaking. Adjustable sensitivity and duration thresholds.
Launch at Login
Start VocaMac automatically when you log in. Always ready when you need it.
Clipboard Preservation
Your clipboard is saved and restored after text injection. Nothing gets overwritten.
Your Voice Stays
on Your Mac.
Period. No cloud. No servers. No analytics. No telemetry. Not even an internet connection required after installation.
Zero Network Calls
Audio is processed 100% on-device using WhisperKit's CoreML engine. Nothing is ever transmitted.
No Audio Storage
Audio buffers are discarded immediately after transcription. Nothing is saved to disk. Ever.
Fully Open Source
Every line of code is public on GitHub. Audit it yourself. AGPL-3.0 licensed. No hidden anything.
No Telemetry
No analytics. No crash reporting. No usage tracking. We literally cannot see what you do.
How we compare
Choose Your Model
From lightning-fast to studio-quality. VocaMac recommends the best one for your Mac.
Get Started in Minutes
Download the signed DMG or build from source. No terminal gymnastics required.
Download & Install
Download the latest DMG, open it, and drag VocaMac to your Applications folder.
Download Latest DMGGrant Permissions & Dictate
Open VocaMac from Applications and grant Microphone, Accessibility, and Input Monitoring permissions when prompted.
Hold Right Option (⌥) → speak → release. Text appears at your cursor. ✨
The Voca Family
Voice dictation, done right, on every platform.
Voice dictation for Linux. X11 & Wayland. whisper.cpp powered.
vocalinux.com →Native macOS menu bar app. WhisperKit + CoreML.
vocamac.comVoice dictation for Windows. Native system tray app. Planned.
vocawin.comQuestions & Answers
Is VocaMac really 100% offline?
Yes. All speech recognition happens locally on your Mac using WhisperKit's CoreML engine. Your voice data never leaves your computer. The only network call is a one-time model download from HuggingFace during setup.
Which Macs are supported?
VocaMac works on any Mac running macOS 13 (Ventura) or later. It runs on both Apple Silicon (M1/M2/M3/M4) and Intel Macs, though Apple Silicon provides significantly better performance thanks to CoreML and Neural Engine acceleration.
How does it inject text into other apps?
VocaMac uses a clipboard-based approach: it saves your clipboard, writes the transcription, simulates Cmd+V to paste, then restores your clipboard. This works in virtually every macOS application.
Why does it need Accessibility permission?
Accessibility permission is required for two things: (1) listening for the global hotkey (Right Option) regardless of which app is focused, and (2) simulating the Cmd+V keystroke to paste text. Without it, hotkeys and text injection won't work.
Can I change the hotkey?
Yes! Open Settings from the menu bar and choose from Right Option, Right Command, Fn, or any function key. You can also switch between push-to-talk and double-tap toggle modes.
How does this compare to Apple's built-in dictation?
Apple's dictation sends audio to Apple's servers for processing (for most languages). VocaMac processes everything locally. Your audio never leaves your Mac. You also get model selection, configurable hotkeys, and open-source transparency.
Is VocaMac free?
Yes, completely free and open source under the AGPL-3.0 license. No premium tiers, no subscriptions, no tracking. Just free software.
How do I update to the latest version?
VocaMac checks for updates automatically once every 24 hours, or you can check manually from Settings → About → Check for Updates. When a new version is available, a banner appears in the menu bar popover. Click it to see release notes, download the signed DMG with progress tracking, and open it for drag-to-replace installation.
What's the difference between WhisperKit and whisper.cpp?
Both use the same OpenAI Whisper models, so transcription quality is identical. WhisperKit is a pure Swift framework optimized for Apple platforms. It uses CoreML and Neural Engine for hardware acceleration, whereas whisper.cpp uses Metal. VocaLinux uses whisper.cpp; VocaMac uses WhisperKit.
Ready to Ditch the Cloud?
Join the early adopters building the future of private, local voice dictation on macOS.