Google App gets Gemini-Input Search Live Feature in AI mode with voice input support

Google On Wednesday, Rol out Search Live in AI mode as part of a small experiment in the US. This feature allows users to discover the back-end-world conversation in real time. Gemini Live But in the Google app. This was first declared as an extension of AI mode in May/O 2025 Kenote session. The search is powered by a custom version of Live Gemini and has voice input capabilities, which can not only have the ability to ask questions, but also follow-up questions.

Live in AI mode

In a blog postIt can be accessed through a wavephor icon below the search bar in the Google app on both Android and iOS devices. Tapping this icon opens a full-screen interface for the search live with two options-and transcript.

The latter enables users to continue interaction by watching a text response and tapping instead of oral command. It will also provide links for relevant websites on the screen, if they want to reach more information.

According to the company, find live tasks in the background and thus, users can leave the Google app and still continue their conversations with search. In addition, they can re -view the conversation by navigating AI mode history. Google says that this feature takes advantage of a query fan-out technique to present users with a variety of useful ingredients from the web.

Currently the facility is being tested as part of the AI ​​mode experiment through opt-in in search labs. It is only available in the US and Google is yet to expand to other areas.

Although the search live in AI mode is currently limited to voice-based command, the company says it will also introduce the ability to indicate the camera on an object and ask questions about it in real time in the coming months.

A previous report It is suggested that this feature supports four different voices – Cosmo, Neso, Terra and Cassini – which can be selected from the three dot menu in the search live interface.