Google has introduced new voice capabilities for the Gemini app on macOS, allowing users to dictate, edit and summarize content directly within the applications where they are working.
Users can activate the feature by pressing and holding the Fn key and speaking naturally into any desktop window. Gemini then converts the speech into polished text at the cursor, removing filler words, recognizing mid-sentence corrections and formatting the transcription automatically.
Users can also enable Gemini reasoning through the app’s settings. With that capability active, Gemini can interpret information visible on the screen and perform more complex tasks.
The system can summarize selected local files, images and documents, compose or rewrite highlighted text and generate or edit images through spoken instructions.
For example, a user could select veterinary records and ask Gemini to summarize a pet’s medical history in an email. A user could also highlight notes and request an executive summary or reference an existing illustration and ask for a modified version.
The feature is designed to keep users within their current workflow rather than requiring them to copy information into a separate AI interface.
Google is rolling out the capability globally to Gemini for macOS users in English, with support for additional languages planned.
