Browse all documentation
Dictating

Dictating into any app

How VoiceSnap Pro gets text into the field you are already using, which apps behave differently, and the focus rules that decide where your words land.

How the text gets in

VoiceSnap Pro does not paste into a window of its own and hope you copy it out. When the transcription comes back, it inserts the characters into the focused field the same way a keyboard would — which is why it works in apps that were never built with dictation in mind, including terminals, Electron apps and web forms.

Because it behaves like a keyboard, everything the field already does still happens: autocomplete fires, an editor's auto-indent runs, a chat box grows to fit, and your app's undo removes what was inserted.

Focus is the whole game

The app that has keyboard focus when you press the shortcut is the app that receives the text. VoiceSnap Pro locks onto it at that moment and names it on the mic pill so you can check before you speak.

Do not click into another window while you are talking. Switching apps mid-dictation is the single most common way to end up with a paragraph in the wrong place — and with push-to-talk you are usually holding a key, so a click is easy to make by accident.

  • A text field must be focused, not merely visible. A page you have scrolled but not clicked into has no cursor.
  • If the pill names the wrong app, release the key, click where you want the text, and start again.
  • Notification banners, Spotlight and system dialogs can steal focus. If one appears while you are speaking, cancel with Escape.

Where it works

Anywhere with a cursor. In practice that covers everything people write in during a working day.

  • Mail clients — Gmail and Outlook on the web, Apple Mail, Outlook desktop, Spark, Thunderbird.
  • Chat — Slack, Teams, Discord, Messages, WhatsApp Web, Telegram.
  • Docs and notes — Google Docs, Word, Notion, Obsidian, Bear, Apple Notes, Confluence.
  • Trackers and CRMs — Jira, Linear, Asana, Trello, HubSpot, Zendesk.
  • Code — VS Code, Cursor, Xcode, JetBrains IDEs, and terminal windows.
  • Anything else with a text box, including browser address bars, search fields and comment boxes.

Where it deliberately does not

Some fields refuse synthetic keystrokes, and that is a feature rather than a bug. VoiceSnap Pro tells you instead of dropping the text silently.

  • Password fields. macOS turns on Secure Input when one is focused, which blocks every app from typing into it — including this one.
  • Elevated windows on Windows, unless VoiceSnap Pro is running elevated too.
  • Some remote-desktop, VM and virtual-machine consoles that grab the keyboard at the driver level.
  • Full-screen games using raw input capture.
NoteAn app can occasionally leave Secure Input switched on after you close a password prompt. If nothing types anywhere on macOS, that is usually why — see Troubleshooting.

Typed insertion versus clipboard insertion

The default is typed insertion, character by character. A few apps drop characters when text arrives faster than a human could type — some Java-based editors, some remote consoles, a handful of older Electron builds.

For those, Settings → General offers clipboard insertion: VoiceSnap Pro puts the text on the clipboard, sends a paste keystroke, and restores whatever your clipboard held before. It is faster for long dictations and more reliable in stubborn apps. The trade-off is that a clipboard manager will record the dictation, and an app that ignores paste keystrokes will not get the text at all. You can set the mode globally or per app.

Dictating in a browser

Rich-text editors — Google Docs, Notion, the Gmail composer — are contenteditable regions rather than plain inputs, and they run their own autocorrect and inline suggestion logic. VoiceSnap Pro types into them normally, but their features still apply, so Smart Compose may complete a phrase you did not say.

If a suggestion keeps merging into your dictation, press Escape after inserting to dismiss it before you keep typing. In Google Docs, leave the browser's own voice typing off — two dictation systems listening at once will double every word.

Habits that make long dictations work

Dictate in thoughts, not documents. One paragraph per dictation gives you a natural checkpoint, keeps each note in the library small enough to find later, and means a mistake costs a sentence rather than a page.

Say “new paragraph” where you want structure instead of reaching for the keyboard, and look at the screen while you speak — the mic pill's level meter tells you whether you are actually being heard before you have spent two minutes talking to a muted microphone.

Still stuck?

If this page did not solve it, send us the app you were dictating into and which of the three checks failed — no mic pill, no level meter, or the wrong app named. That is usually enough to answer in one reply.

Back to documentation