Most dictation tools transcribe into their own window and leave you to copy the result across. Dictami doesn't have a window to copy from — it types into whatever app you were already working in.
A transcription window sounds harmless until you count the steps. Speak, wait, read, select, copy, switch back, find the cursor, paste, fix the spacing. That's eight actions around one sentence, and every one of them is a chance to lose your place in what you were writing.
The whole value of dictation is that speaking is faster than typing. Spend that saving on window management and there's nothing left.
Dictami inserts text through the same macOS accessibility interface that assistive software uses — the layer designed for exactly this. As far as the receiving app is concerned, the words arrived the way words normally arrive. Nothing has to be built for Dictami, and no app needs a plugin.
This is why macOS asks for Accessibility permission on first launch. It's not analytics and not monitoring: without it, Dictami could hear you but would have no way to put the result anywhere except its own window.
Native Mac apps — Mail, Notes, Messages, Pages, Reminders. Browsers and anything running inside one, including Gmail, Google Docs, and web-based chat. Electron apps like Slack, Notion, Discord, and Obsidian. Editors and IDEs: VS Code, Xcode, JetBrains, Sublime. Terminal windows. Search fields, comment boxes, form inputs, and the address bar.
If you can put a cursor in it and type, it works.
Secure input fields are the honest exception. When macOS is in secure input mode — a password box, a login prompt, the keychain — the system blocks every application from putting characters in, and Dictami is no exception. That block exists to stop keyloggers, and it's doing its job.
A handful of games and remote desktop clients read the keyboard at a lower level and won't see inserted text either. Everything you'd actually write prose into is covered.
There's nothing to configure per application. The same key works in every app — hold it, speak, release. The default is the right Command key, and it's captured cleanly so no other app reacts to it while you dictate.
macOS 14 Sonoma or later on an Apple Silicon Mac — an M1 or newer. Intel Macs aren't supported. Recognition runs on your Mac in 30 languages, with punctuation, and needs no internet connection after the first download.
7-day trial inside the app. No card, no account.
Download Dictami