Dictate into any app on your Mac.
Hold a shortcut and speak. A live transcript pill shows every word; let go, and your words land in the text field you were typing in: mail, chat, documents, your code editor, anywhere.
One subscription. Your Mac too.
In VoiceType on your iPhone, sign in with a Google or email account and subscribe to VoiceType Monthly. Then sign in to VoiceInput on the Mac with that same account. The Mac unlocks automatically; there is nothing to buy on the Mac.
Everything you need to talk, not type.
Built for people who write all day: fast to start, honest about what it heard, and careful with your words.
Any app, any text field
VoiceInput types where your cursor is, in every app. It never types into a password field.
Hold to talk, or go hands-free
Hold ⌃⇧Space while you speak, or press ⌥Space to talk hands-free and press it again, or any key, to finish.
A live pill that never steals focus
A floating transcript shows every word and a live mic level, so you know you are heard. Move it, resize it or hide it.
Your vocabulary, your spelling
Names, products and acronyms come out your way, with the same vocabulary as your phone.
Clean-up, one switch at a time
User prompt, obvious grammar, custom vocabulary and tone are separate, visible switches. Clean-up tidies what you said and never answers it.
Tones, including code
Formal, Casual, Coding or your own. Press ⌥J and describe the code you want: you get real Java, pasted with its line breaks.
Knows what is on screen
What is on screen goes with a dictation to the clean-up model, so names and terms match. On by default, never from a password field, and you can switch it off.
Never loses a word
Every transcript is copied to your clipboard and saved to your dictation history, with one-click copy.
Silent, and never presses Return
No chimes or spoken alerts, ever. Line breaks become spaces, so a chat message is never sent halfway. You press Return.
Keeps going offline
If the network drops, the Mac's built-in recognizer takes over, so a dictation is not just lost. On macOS 14 and 15 that works offline once macOS has its on-device speech model.
Follows your microphone
Plug in a headset or dock your laptop and VoiceInput uses the microphone you are actually speaking into.
Updates itself quietly
New versions install in the background without a dialog and never interrupt a dictation.
Capture it with one shortcut.
Record a meeting, grab a screenshot, or hear a page read back, and turn it into a task without breaking your flow. Tasks go to your account's to-do list, the same list the To-Do List app shows; ⌃⌥T shows today's.
- Recordings and live meeting notes with speakers labelled
- A screenshot of the screen or a region, turned into a task
- Read the selected text aloud, or add it as a task
Every shortcut, at a glance.
These are the defaults. Change any of them in Settings › Shortcuts.
| Shortcut | What it does |
|---|---|
| ⌃⇧Space | Hold to talk; let go to type it in |
| ⌥Space | Hands-free: press to start, press again (or any key) to finish |
| ⌥J | Java code: describe or dictate code and get real Java |
| ⌃⌥T | Show today's tasks |
| ⌥⇧1 | Record audio, speakers labelled |
| ⌥⇧7 | Record a single speaker |
| ⌥⇧2 | Live meeting notes |
| ⌥⇧3 ⌥⇧4 | Screenshot of the screen, or of a region, as a task |
| ⌥⇧5 | Read the selected text aloud |
| ⌥⇧6 | Add the selected text as a task |
Up and running in two minutes.
- Download and installOpen VoiceInput.dmg and drag VoiceInput into Applications. Then open it: a microphone appears in your menu bar.
- Allow two permissionsMicrophone, so it can hear you, and Accessibility, so it can type into other apps. Screen & System Audio Recording is optional, for screen context and recordings.
- Sign in and talkSign in with Google or email: the same account you signed in with in VoiceType on your iPhone, where you subscribed. Then hold ⌃⇧Space and speak.
Requires macOS 14 Sonoma or later on Apple silicon or Intel. Cloud speech recognition needs an internet connection.
Mac questions, answered.
Is VoiceInput for Mac free?
The download is free. Using it needs VoiceType Monthly, which you start in the VoiceType app on iPhone ($4.99 a month, with a free first week for new subscribers). Sign in to VoiceType on the iPhone with a Google or email account, then sign in on the Mac with that same account and it unlocks. A subscription with no Google or email sign-in on the iPhone cannot unlock the Mac. There is nothing to buy on the Mac.
Why does it need Accessibility permission?
macOS only lets an app type into other apps once you allow it under Privacy & Security › Accessibility. VoiceInput uses it to put your words where your cursor is, and never types into a password field.
Does it type while I am still talking?
No. The live pill shows your words as you speak; nothing goes into your app until you finish, and then it all goes in at once. The full transcript is also copied to your clipboard.
What does screen context read, and can I turn it off?
It is on by default. Whenever a clean-up step runs, VoiceInput reads the app's name, the window or tab title, the text around your cursor and the text visible in the window, and sends it with that dictation to the clean-up AI model, so names and terms match what is on screen. It never reads a password field and VoiceInput does not keep it. Turn it off in Settings › Post-processing; the ⌥J Java code shortcut always reads the screen.
Does it work with my microphone or headset?
Yes. VoiceInput picks the microphone at the start of every dictation and follows changes, so a newly connected headset or dock microphone is used right away.
How do updates work?
VoiceInput checks for updates and installs them quietly in the background. You never need to download it again.
Where do my voice and text go?
While you dictate, your voice is streamed to the speech engine picked in Settings (ElevenLabs by default), which sends back the text; VoiceInput keeps no audio from dictation. With a clean-up step on, the text and the screen context go to the clean-up model (Qwen on Cerebras by default). Pick Built-in dictation in Settings to use the Mac's own recognizer. The details, and every provider, are in the VoiceInput privacy policy.
