Google has updated Gemini for Mac to version 1.88, adding voice control capabilities that were previewed at Google I/O 2026 in May and promised as a summer update. Starting today, users can long-press the Fn key from any window to speak naturally to Gemini, with the app handling transcription, filler-word removal, and, optionally, acting on whatever is visible on screen.
The feature ships in two distinct modes, intelligent dictation listens to your speech and enters clean, formatted text at your current cursor position, automatically stripping “umm” and “ah” while accounting for mid-sentence corrections so the final output reflects what you meant rather than exactly what you said. The second mode, screen-aware reasoning, is opt-in and allows Gemini to see what is on the Mac’s display, letting users highlight files, images, text, or documents and issue a voice command to have Gemini edit, summarize, or compose content directly inside the app they are working in.
What the Fn key shortcut actually does
The choice of Fn as the trigger is notable because macOS already uses a long-press on Fn to invoke the system’s own dictation feature. Google is planting a flag on the same key with a different experience, one that is AI-aware rather than a straight speech-to-text transcription pipe. Google positions this alongside Gboard Rambler, its upcoming voice-input feature for Gemini Intelligence phones. That this is part of a broader cross-platform voice strategy rather than a Mac-specific experiment.
Google demonstrated the potential at I/O 2026 with a scenario showing a Mac user selecting files and an image in Finder, then issuing a single voice command via Gemini to draft and send an email to a dog kennel with the relevant information and photo attached. That kind of multi-step file-and-app action goes well beyond what standard macOS dictation can do.
Where this fits in a busy summer for the Mac app
Voice control is the third significant update Gemini for Mac has received since early July. On July 1, Google launched Gemini Spark for the Mac app, giving the AI agent the ability to act on locally stored files rather than just responding inside a chat window. Spark remains in beta and is limited to Google AI Ultra subscribers aged 18 and over in the United States, with Google AI Ultra starting at $99 per month. Then on July 21, version 1.86 brought the Neural Expressive redesign, aligning the Mac app visually with the Android, iOS, and web versions. Today’s voice rollout completes what Google effectively announced as a summer update package.
The Gemini Mac app only launched in April this year, built as a 100% native Swift app by a small team that shipped more than 100 features in under 100 days. At launch, Gemini was the last of the three major AI services to arrive on Mac, coming after both ChatGPT and Claude had established dedicated apps. Four months in, it has closed a meaningful portion of that gap in terms of feature depth.
Voice control is rolling out globally to all users in English, with additional languages listed as coming soon. The app requires macOS 15 or later and is free to download from gemini.google/mac, though Gemini Spark and some agentic features remain behind the AI Ultra subscription tier.
