- Google announces new voice features rolling out for the “Gemini” app on macOS (Gemini on macOS).
- A voice transcription feature that transcribes, edits, and summarizes the user’s voice.
- Intelligent speech recognition is enabled by default, performing real-time tasks such as automatically removing filler words like “um” and “uh” and correcting sentences.
On Wednesday, July 29, 2026 local time, Google announced new voice features rolling out for the “Gemini” app for macOS (Gemini on macOS), which was originally released in April 2026.
The new voice feature rolling out to the Gemini app for macOS is a voice transcription function that transcribes, edits, and summarizes the user’s speech. Intelligent speech recognition is enabled by default, automatically removing filler words like “um” and “uh” and correcting sentences in real time.
It is easy to use—simply talk to the voice input tool activated by holding down the Fn key. Notably, this voice transcription feature, including the automatic removal of filler words, is expected to be an adaptation of “Google AI Edge Eloquent,” an AI voice transcription app powered by the “Gemma” small language model that was previously released for iOS devices.
Furthermore, going beyond simple voice transcription, enabling “Reasoning” in the settings allows Gemini to understand the context on your screen and handle more complex tasks. For example, with a file or image on your desktop selected, you can speak commands like, “Read this vet file and compile my dog’s medical history into an email to the kennel,” and it will extract, summarize, and convert the content into text.
Additionally, by simply selecting any text on the screen and saying, “Turn this memo into a summary with a TL;DR at the beginning,” you can rewrite text and adjust its tone. Alternatively, by selecting an image and instructing, “Make this illustration dark mode,” it can also generate and edit images.
The newly announced voice features for the macOS version of the Gemini app are rolling out globally, starting first for English-language environments. Support for additional language environments is planned to expand in the future.
Source: Google (https://blog.google/innovation-and-ai/products/gemini-app/speak-naturally-gemini-app-mac-os/)