Support

Microphone, to hear you while you hold the key. Accessibility, to notice the key you hold anywhere on the Mac and to type the finished text into the app you're using. TypeVoice never reads what is on your screen. Both switches live in System Settings › Privacy & Security.

System Settings › Keyboard › "Press 🌐 key to" → Do Nothing. Or pick another trigger under Settings › Dictating › Custom and press the key or combination you want: Right ⌘, Right ⌥, ⌥Space, anything.

Flip the TypeVoice switch off and on in System Settings › Privacy & Security › Accessibility, then quit and reopen the app. The menu bar shows a fix-it item whenever the permission is missing.

Some places can't take text (the desktop, a button, a list). The pill offers a Copy button in that case, and every dictation is kept in Summary where you can copy it later.

The app is signed with a Developer ID and notarized by Apple, so a normal download opens with no warning. If macOS still objects, the download was probably interrupted: delete it and download again from this site. Never disable Gatekeeper for it.

Privacy › Reveal in Finder opens the folder. Summary is an SQLite database, the dictionary is JSON, and recordings (if you turned on Offer the audio after each dictation) are small AAC files. Summary › Export writes everything as text, Markdown, CSV or JSON. Delete Everything removes all of it.

Settings › Indicator › Offer the audio after each dictation. From then on a Drag audio chip appears on the pill after you speak, and every row in Summary can be played or dragged. Drop it into iMessage, WhatsApp, Slack or an email. Recordings are AAC, about 240 KB a minute, with long pauses trimmed so the note stays tight (the words are never touched); they stay on your Mac and are deleted after a week unless you choose a longer retention (or forever) next to that setting.

Yes. Open the URL typevoice://toggle to start or stop a hands-free session; typevoice://start, typevoice://stop and typevoice://cancel do one thing each. In Shortcuts use the Open URLs action; in a terminal, open typevoice://toggle.

Yes. After the one-time model download on first launch, dictation needs no connection at all. A licensed app re-checks its key once a week when it can; a Mac that stays offline keeps working for 30 days before it needs to.

English, all accents. The recognition model is English-only by design; it is what makes it this accurate and this fast.

Buy once from the pricing section; Dodo Payments handles the checkout and emails you a license key straight away (the thank-you page can also hand it to the app directly). Paste the key into TypeVoice › License. A personal key works on 2 Macs at once (a desk Mac and a laptop); to move to a new Mac, press Deactivate this Mac on the old one first, or email us and we'll free the seat. Teams: one key for 5 people, two Macs each — see pricing.

The site shows the US price, $79, and the checkout charges it in your own currency, plus tax where it applies. Five countries have a lower price set for local purchasing power: ₹2,499 in India, and $24 in Pakistan, Bangladesh, Egypt and Nigeria (the team key ₹9,499 and $89). It goes by the billing country you enter at checkout. Same app, same license, same updates.

Search your email for "Dodo Payments". If it's gone, write to support from the address you bought with and we'll resend it.

Within 14 days of purchase, email support and we refund in full, no questions asked. Dodo Payments returns the money to the original payment method, usually within a few business days.

The app checks for updates once a day and offers them in place; Settings › General has Check for Updates and a switch for the automatic check. Updates are signed twice: by Apple's notarization and by our own update key.

macOS 26 or later on Apple silicon. The speech model runs on the Neural Engine, which is why it is this fast and this light on the battery.