Google Tests Audio Overviews for Search Results Powered by Gemini

Google is experimenting with a new feature called "Audio Summaries," which provides brief AI-generated audio overviews of search results. Available through Search Labs, this feature is currently limited to users in the United States (and only supports English). Clearly designed to facilitate information comprehension through listening rather than reading, the functionality is powered by Gemini, the company's advanced AI model that also supports products like NotebookLM.

Under this feature, for certain topics (such as explanations of how something works or general inquiries), the company now offers a button labeled "Generate Audio Overview." Once clicked, the system creates a concise audio summary in about 40 seconds. The information in each audio overview is derived from existing web content, and Google provides links to the original sources below the player for users who wish to read more.

What sets this feature apart is the way the audio is presented. Unlike typical robotic voice assistants, the audio takes the form of a dialogue between two AI-generated voices (resembling casual discussions or podcasts). These summaries typically last around four minutes but may vary depending on the topic.

Additionally, the audio player includes basic controls such as play, pause, volume adjustment, and speed settings (ranging from 0.25x to 2x). Users can rate the summaries by liking or disliking them to help improve the feature. Currently, audio overviews are only available to those who opt into Search Labs, accessible via the Google app or Chrome browser. However, the tech giant has not yet announced when it will roll out the feature to a broader audience.

The latest addition, audio overviews, is part of a broader effort to make search results more interactive and easier to use in various situations. Earlier, at its I/O 2024 conference, the tech giant unveiled several new AI-powered updates aimed at enhancing the search experience, including "AI Overviews" and later the new "AI Mode." Through these innovations, the company is working towards making information more accessible by integrating text, images, and now audio.

However, Google’s AI endeavors have not been without controversy. One major issue involves AI Overviews, which some publishers and websites claim have led to a drop in traffic. This feature was even temporarily suspended in 2024 for providing inaccurate responses, such as suggesting users apply glue on pizza to keep cheese from falling off. More recently, the company paused the rollout of its AI-driven "Ask Photos" feature (which uses a dedicated version of the Gemini model) due to concerns about latency and quality.