Gemini for macOS adds intelligent dictation that cleans up speech in real time

Google added a system-wide dictation mode to the Gemini app for macOS that transcribes spoken words into any desktop window while automatically stripping filler words and fixing mid-sentence corrections.

What's new

The feature lets a user speak into any application on the Mac — not just the Gemini app itself — and have that speech converted to clean, formatted text at the cursor position. Per Google's announcement: "This intelligent dictation feature allows you to speak naturally into any window on your desktop, transcribing your spoken words into clean, polished text." The system also handles the messiness of natural speech: "It automatically removes the 'ums' and 'ahs,' catches mid-sentence corrections, and drops the formatted text instantly at your cursor."

Context

This lands alongside a broader push by Google to make the Gemini app a system-level utility on macOS rather than a standalone chat window, following earlier moves like AI Edge Gallery and on-device dictation tooling. Voice-to-text has become a competitive front among AI assistants — OpenAI, Anthropic, and Google have all shipped increasingly capable dictation and voice-input features this year as a way to reduce friction for users who'd rather talk than type, particularly for drafting long-form text like emails or code comments.

Why it matters

System-wide dictation that cleans up disfluencies automatically is a small feature on its face, but it targets a real workflow gap: most native macOS dictation tools transcribe literally, leaving in every stutter and false start. By handling cleanup client-side across any app, Google is positioning Gemini as an OS-level utility rather than a destination app — a strategy that increases daily engagement surface area without requiring a user to open a chat window at all.