Google adds voice control to Gemini for Mac

Google has introduced a new voice-driven AI feature for its Gemini app on macOS, allowing users to interact with the assistant from any desktop window by holding down the Fn key. The update, which began rolling out globally on July 29, 2026, includes an enhanced dictation mode and an optional screen-aware reasoning capability, positioning Gemini as a more integrated desktop competitor.

The core functionality is activated by a long press of the Fn key, which summons Gemini for voice input directly within the active application window. By default, this enables "intelligent dictation," a feature designed to transcribe spoken words into polished text. Google states the system automatically removes verbal fillers like "ums" and "ahs," catches mid-sentence corrections, and inserts the formatted text at the user's cursor without requiring them to switch apps.

For more complex tasks, users can opt into "screen-aware reasoning." This optional mode grants Gemini permission to analyze on-screen content to fulfill context-dependent requests. For example, a user could highlight notes in a document and ask Gemini to turn them into an executive summary, with the revised text placed directly...