Gemini for macOS has just added a more natural way to work: you can now dictate, summarize, rewrite text, and generate images using your voice, without leaving the window you’re already working in. The feature aims to reduce constant switching between apps and help you stay focused on your workflow.
Smart dictation with one key
By holding down the Fn key, you can talk to Gemini from any desktop window. By default, the app activates smart dictation, which turns your words into clean, properly formatted text.
The tool removes filler words such as “uh” or “um,” recognizes corrections made halfway through a sentence, and places the result directly wherever your cursor is. It’s similar to dictating a message, but without having to manually clean up every pause or mistake.
So what can you use it for? You can draft an email while reviewing a document, take notes during a meeting, or write down a quick idea without taking your hands away from your main task.
Gemini understands the context on your screen
If you want Gemini to do more than transcribe, you can activate the reasoning feature in the settings. With this option, the app can interpret the content you have selected on screen and help you carry out more complex tasks.
Some of the announced possibilities include:
- Extracting and summarizing information: select local files, images, or documents and ask for a specific result. For example, you can request a summary of several veterinary files to prepare an email about your pet’s medical history.
- Drafting and rewriting text: select any text and ask Gemini to change its tone, improve the writing, or turn a few notes into an executive summary with the main points at the beginning.
- Generating and editing images: describe a new image or use an illustration you already have on your desktop to request variations, such as a version adapted for dark mode.
The important difference is that you don’t have to copy and paste the content into another app. Gemini uses what you select and places the result where you need it, as long as you grant the corresponding permissions.
AI doesn’t replace your judgment: it speeds up repetitive tasks so you can focus on reviewing and deciding.
Feature availability
The new voice capability is rolling out globally to users of the Gemini app on macOS, although it currently works in English. Google says it will add more languages soon.
You can download the app from gemini.google/mac. As with any tool that analyzes files or the content on your screen, it’s a good idea to review your privacy settings and activate reasoning only when you actually need it.
Talking to an app to write, summarize, or transform an image is no longer a futuristic demonstration. In this case, the proposal is much more everyday: press a key, explain what you need, and receive an integrated response within your workflow.