Google’s latest update to its AI-powered research tool NotebookLM can turn the materials you want to deliver into an audio discussion like a podcast. a new feature called Audio Review takes data from the documents you upload and then creates a “deep dive” discussion between two AI hosts. In addition to summarizing your sources, Google says hosts will be able to find links between different topics and even banter back and forth. Based on the sample the company posted with the announcement, the AI owners sounded human enough to listen, though you can still tell the AI-generated voices from the inflection and odd pronunciations of certain words.
As the feature is still in the experimental stage, Google admits that it has its limitations. The hosts can only speak English at the moment and they sometimes give inaccurate information, which means you need to double check your material and make sure you’re not just learning something that isn’t factual. You also can’t interrupt hosts while they’re still talking, and NotebookLM still takes a few minutes to generate an Audio Preview for notebooks with larger files. Google Labs product manager Biao Wang wrote in the feature’s announcement that his team was “excited to bring voices to NotebookLM” despite these limitations because they “know that some people learn and remember better by listening to conversations.”
The company launched NotebookLM in 2023 as a kind of digital assistant that you can ask questions about the documents you download. In June of this year, Google announced the official launch of NotebookLM It runs on Gemini 1.5 Prointroduced it with new features and tools and expanded to more than 200 countries and territories.
