Search Live: Voice Conversations with AI Mode

Google has officially launched Search Live, a pioneering feature that allows users to engage in voice conversations with the search engine through AI Mode. Currently available in the United States for users participating in the AI Mode experiment on the Google app for Android and iOS, this innovative approach aims to enhance the way users search for information.

With Search Live, users can initiate a dialogue by simply tapping the new “Live” icon in the Google app and verbally asking questions. For example, if you’re preparing for a trip, you might ask, “What are some tips for preventing a linen dress from wrinkling in a suitcase?” In response, the AI generates helpful audio answers and provides users with links to delve deeper into the topic.

This feature is particularly advantageous for individuals who are multitasking. Since Search Live operates in the background, users can continue their voice conversation even while using other apps. Additionally, for those who prefer text-based responses, tapping on the “transcript” button allows for viewing the text reply, and follow-up questions can be typed in for further clarification.

Search Live is powered by a customized version of Gemini, incorporating advanced voice recognition capabilities. This custom model ensures that whether users are verbally querying or typing their questions, they receive reliable and informative responses. Furthermore, Search Live utilizes a technique known as query fan-out to broaden the range of helpful web content displayed, opening avenues for more diverse exploration.

The accessibility of Search Live is straightforward, with an easily identifiable icon located directly beneath the search bar for those who opt into the AI Mode experiment. Users are encouraged to test this feature and share their feedback on their experiences.

Looking ahead, Google anticipates expanding the capabilities of Live in AI Mode, including future updates that will allow users to engage using their camera. This will enable users to show the search system what they’re observing in real time while conversing.