On July 29, 2026, Google introduced an updated voice shortcut for its Gemini app, specifically for macOS users. The new function lets people press and hold the function (fn) key to use voice commands from within any open application. This means users can speak their instructions directly to the AI without the need to navigate away from their current window or app.
The latest addition does not stop at voice recognition. After receiving a command, Gemini can analyze what is visible on the screen. Users can ask the AI to revise parts of a document, summarize notes, or reword a sentence — all done on the spot. This new capability eliminates the need to copy and paste content into another tool, making workflows significantly more efficient.
Privacy built in — with a toggle
Google is aware that some users may feel uneasy about an AI analyzing their screen activity. To address this, screen-aware reasoning is not activated by default. Users are given the choice to enable the feature or keep it off entirely. This ensures that Gemini can operate with voice commands without monitoring or interacting with content displayed on the user's screen if the user so chooses.
Smarter transcription for cleaner output
In addition to the voice shortcut, the voice dictation system in Gemini has been enhanced to produce more refined text. The AI automatically filters out common filler words such as “um” and “ah,” transforming spoken input into clean, polished writing. This ensures that the final output is more professional and easier to integrate into emails, reports, or documents without significant editing.
The new voice capabilities are now available globally to all macOS users of the Gemini app in English. Google has confirmed that additional language support will be added later in 2026, making the feature accessible to more users around the world.
With these updates, the Gemini app is moving beyond being just a chatbot. It is becoming an integrated tool that supports hands-free, efficient workflows. By combining voice commands with screen context and smart transcription, Google is helping users work more intuitively without the need to juggle multiple applications. Gemini allows users to extract and summarize information from local files, rewrite text, and even generate and edit images using voice commands. For example, users can ask Gemini to summarize their dog’s medical history or turn notes into an executive summary. These voice features are activated by long-pressing the Fn key or using the new screen sharing button in the “Ask Gemini” prompt box. Gemini will show a floating pill with a waveform at the bottom of the screen, and users must have the latest version (1.88) of Gemini for macOS to access these new capabilities.