Google introduces natural speak and voice typing features for Gemini on macOS

Introduction to Gemini Voice Typing on macOS

Google has introduced a new suite of natural language and voice typing features for the Gemini app on macOS, designed to keep users within their active workflows. Rather than requiring copy-and-paste actions between separate chat interfaces, the update allows desktop users to create, edit, and summarize content instantly using their voice.

Core Dictation and Transcription

By default, long-pressing the Fn key activates intelligent dictation inside any window on the desktop. The system is designed to transcribe spoken words into polished text while automatically eliminating filler words like "ums" and "ahs." Furthermore, it accommodates mid-sentence corrections and places the formatted output directly at the cursor location.

Advanced Reasoning and Screen Context

For users who want to move beyond basic dictation, optional settings enable Gemini reasoning to analyze active screen context and assist with complex desktop tasks.

• Information Extraction: Users can highlight local files, images, or documents and issue voice prompts to summarize data, such as compiling pet medical records into an email.
• Text Composition: Selected text anywhere on the screen can be rewritten or restyled using voice commands to generate executive summaries or adjusted tones.
• Image Generation: Visual concepts and existing illustrations can be iterated upon or transformed into alternate formats, such as dark-mode views, through spoken instructions.