Google's NotebookLM: Transforming Research into Interactive AI-Generated Podcasts

I write about AI for a living -- and NotebookLM is the most exciting tech to arrive since ChatGPT

At first glance, Google's new NotebookLM looks like just another generic AI application. For those not in the know, it's a free service where you can upload a variety of media, and query this content using natural language, rather like using ChatGPT to summarize a business report. So far so ho-hum, right? Wrong. This is where most people miss the really incredible value this tool will bring to the future. I call it 'unlocked information'. The problem is the world is awash with data. So much data. It's coming at us from everywhere. The web, YouTube, television and radio, our phones. We're literally drowning in the stuff. And nowhere more so than in the scientific research field. There are thousands of new scientific discoveries happening every single day, all of which have some sort of research paper or extra information attached to them. But how can we mere mortals find out about all this progress? Even if we know where to look, the information that's out there will generally stump the layperson because of dense science, complex math or deeply confusing concepts. That's where NotebookLM comes into play. The beauty of NotebookLM is you can throw all sorts of information into the box - videos, audio, research papers, newspaper articles, even links to transcripts of TV programs. The AI will then 'read' the material, translate and make sense of it, and spit it back out in the form of a 5 to 10-minute podcast featuring two chirpy personalities. It's hard to explain how useful and compelling this can be. I recently gave my wife the link to the Google product and she immediately dropped in some information on photobiomodulation, an incredibly niche offshoot of functional medicine that is slowly gaining ground as a pathway to well-being. Within 10 minutes she had an informative, and most importantly, 'interesting' podcast she could listen to on her phone to get an idea of what the science was about, and whether it was important to know more about the subject. Now she could have gained the same knowledge by going through the paper itself in detail, assuming she could understand all the medical terminology and jargon, but it would have forced her to devote a fair amount of time and effort to something that may have turned out to be a dead end. By delivering the information via two people chatting around a 'virtual desk', it suddenly became so much more accessible. Of course, this sort of AI tool is not fully formed right now, but this is just the beginning. It's not hard to imagine a near-distant future where we can all access important and meaningful information in a way that is not dry and academic, but which comes alive at the touch of a button. It's starting with audio podcasts, but soon it will be able to generate human-like AI personalities, which, like newsreaders and documentary presenters, will deliver personalized video infotainment-on-demand in ways that make it fun to learn, not a drag. Why should 90% of the world's population be excluded from learning about their world and things that matter, just because they don't have the patience or desire to wade through treacle? And in case anyone things it's all about lofty, esoteric information only, just consider a world where we can do a search on a recipe or a DIY repair procedure, and instead of watching a canned YouTube video or a static web page created two years ago, we will receive a personalized response built around what the AI already knows about our needs and environment. How much more effective for it to know you're lactose intolerant, or you don't like salt when delivering a dinner recipe for an upcoming birthday treat? This is just the tip of the iceberg, but it should be no surprise that it's search king Google leading the way with this sort of transformative tech. There's a reason podcasts are booming right now, and it's because they are entertaining as well as informative. It's like eavesdropping on a chat over the garden fence, we get involved whether we want to or not. For the doubters out there, I offer a simple challenge. Think of some interest you have where you'd like to learn more without much effort. It can be anything from cooking to rocket science. Then do a quick search to locate some scientific or other relevant information on the subject. It can be a video, audio, PDF file or even website URL. Then go to notebooklm.google.com and upload your source data and press the Generate button on the podcast tab. It's kind of hidden, but if you look hard enough you'll find it. Once the audio generation has completed in around 5 minutes or less, listen to the results with an open mind. Don't just rate it now - yes the voices can be annoying, and the delivery is also sometimes a bit too chippy - but consider what it might be like to have that kind of search access in every aspect of your life. Not just audio either, but video, chat, on the phone, everywhere. It's now pretty clear that ChatGPT was just the bare minimum tech out of the gate. Not fully formed, and not even a fraction of what we've got coming. If you want to know what the future looks like, Notebook LM is a great signpost on how this new AI age will integrate with all of our lives, every day, all the time.

I turned my book into a podcast using Google NotebookLM's updated feature -- here's how

Here's how to summarize YouTube videos and create shareable AI-generated discussions. In a new round of updates for NotebookLM, Google announced yesterday (September 26) that the model is now capable of summarizing YouTube videos and creating shareable AI-generated discussions. It really seems like there is no end to what AI is capable of, especially for creatives. That's why I couldn't wait to see what Google's Notebook LM, powered by the company's multimodal large language model Gemini 1.5 Pro, would do when I shared my young adult science fiction book with it. Earlier this year, I wrote a book for my Fortnite-loving son after he told me he couldn't find anything good in the library. Of course he was being overly dramatic, but since I already had a book idea, I got to work. My kids and their classmates enjoyed the book, and it turned out to be a best-seller on Amazon (for a week-ish, but I'll take it considering I only wrote it for my son). I wasn't planning on doing anything else with the book until I discovered Google NotebookLM. Utilizing Google Gemini AI, the model is designed for AI-powered note-taking and has AI research tools and audio capabilities. Google's NotebookLM recently expanded source options and enhanced citation features for more accurate research. The announcement of these key updates means it's a great fit for writers and anyone looking for deeper interactions with their data in a safe, efficient, and reliable manner. One particular feature that stood out to me is NotebookLM's podcast-style summaries. The audio overview capability of NotebookLM, enables users to create podcast-style summaries from their selected sources. Essentially, the model can take anything from an article or blog post to an entire book manuscript and produce an audio file that sounds exactly like two people naturally engaged in a discussion about the content. This is an innovative addition to the model's capabilities, as now, instead of reading through texts, users can listen to an audio version of a customized study guide or summary. With the latest update, the AI can even summarize YouTube videos. This makes it a perfect feature for multitasking; that way, one could digest information on the go, just like listening to a podcast or an audiobook. This function is a godsend for anyone who has an extremely busy schedule. For example, if my son did not want to read my book, he could listen to the AI-generated podcast of it on the way to soccer practice. But, please don't tell him that. The AI-generated podcast of my book is so convincing, you'd instantly want to subscribe if it were real. The way the two voices engage with each other and elaborate about the content with scoffs, tiny ums, and laughter is so natural and human-like it'll make you question reality. Still, Google is quick to point out that NotebookLM is experimental. Like most AI models, it's not beyond making errors or inaccuracies from time to time. However, the updates this month have radically improved its usefulness, and NotebookLM is better positioned than ever before to cater to user demands for an effortless, AI-driven information management system. One important note is that NotebookLM users have to be at least 18 years old, have a Google account, and live in one of the 180 regions where the Gemini API is available. Access to NotebookLM is free; users are not charged, and according to Google, your data is not used to continue training the AI, which is a key privacy consideration in today's AI landscape.

This Google AI Tool Can Turn Your Research Into a 'Podcast'

Turn your documents into audio summaries, thanks to Google. Credit: Google When it comes to Google's AI tools, you might think first about the Gemini assistant app, or maybe some of the cool tricks you can do in Google Photos. However, Google has another powerful artificial intelligence tool that's great for collecting and collating notes on any kind of topic. It's called NotebookLM. The aim of NotebookLM (the LM stands for Language Model, of course) is to help you make sense of whatever it is you're researching. It'll lend a hand by creating summaries, answering questions about the documents you've gathered, and linking together data points where required. As is normal with this type of AI, it's still considered an experimental feature, and you need to double-check everything it produces for hallucinations. That said, it's a potentially useful tool, and just got another new feature: Audio Overviews. Essentially, this feature can create short podcasts, with AI podcast hosts that summarize all the information you've brought together. Whether you're researching 19th century literature or which new phone to buy, the Audio Overview will present it in a friendly, accessible way -- and the results can be pretty impressive. You can access NotebookLM, and the Gemini models underpinning it, for free using your Google account. Give it a try with your own research notes, and be prepared to be amazed at how natural the resulting clips sound. If you're completely new to NotebookLM, you can sign in with your Google account to get started. You'll see some example projects (called notebooks) included for you to have a play around with. Open the Invention Of The Lightbulb one, for example, and try asking the AI questions about the work of Joseph Swan, Thomas Edison, and others. The question-and-answer routine will be familiar if you've already used something like ChatGPT, Gemini, or Copilot -- but here, the training data used for the model is only the sources you've uploaded to your notebook, rather than the wider web. You can point NotebookLM to web links though, as well as add plain text, PDFs, Google Docs, and Google Slides (each notebook can have up to 50 sources). To start putting together your own notebook, click the NotebookLM heading (top left), then select New notebook. You'll be asked to start adding sources to your new notebook; you can upload documents from your computer, pick them out from Google Drive, point NotebookLM towards web links, and paste in text from the clipboard. You can add more sources later by clicking the + (plus) button next to Sources on the left-hand panel. As soon as you start adding sources, NotebookLM will start summarizing them for you, then suggest questions you might want to ask. You can click individual sources to view them separately, and follow the onscreen links to put together a FAQ, a study guide, a table of contents, a timeline, or a briefing doc for everything you've collected. For the purposes of this guide, I uploaded a PDF of s recent study on how smartphone apps could be developed to measure blood pressure, with no cuff required. I also added the accompanying press release as a web link, to give NotebookLM a bit more material to work with. Right away the AI suggested question prompts around how the app utilizes existing smartphone technology, what the potential benefits might be, and what the challenges are in developing this tech -- and on the whole, NotebookLM came up with responses that made sense and that were accurate based on the material provided. (As you use the app, the answers you get come with citations that take you back to the sources of that information.) To produce your podcast (or Audio Overview), you need to open the Notebook guide (there'll be a link on the right if you're on the chat screen), then click Generate on the right. As of now, there are no settings you can play around with -- you get the same two podcast hosts every time, one male voice and one female voice, talking in English. After a few minutes of thinking, the audio starts playing, and you can download the file too if needed. In my test, the Audio Overview did a fine job of summing up the blood pressure app research with a high level of accuracy, producing what sounds like a natural podcast conversation between two people -- though it did miss some nuance, and was a little generic in places. As a tech demo and a summarization tool it's really impressive, but if details truly matter in your work, I wouldn't rely on AI entirely at this point.

Google's scary-good AI podcast tool just got better with built-in YouTube upload and easy sharing

Google's NotebookLM AI tool now works with YouTube video URLs and audio files as sources to help you create engaging audio podcast discussions with just one click. You can also share your creations easily with a public URL for your podcast. Just a few weeks ago, Google introduced NotebookLM and TechRadar's AI Senior Editor has been experimenting with the tool that lets you quickly convert documents into incredibly realistic conversations between two voices. Graham wrote, 'I've had a glimpse of how AI will change the world, and it's filling me with both fear and wonder in equal measure.' Now, Google has expanded the AI tool with new source types including the ability to upload YouTube videos as well as audio files. These new formats will allow users to take full advantage of NotebookLM to analyze video essays and lectures, make group projects far easier by letting you summarize the audio from a group call, and quickly turn class recordings into study guides. In its original version NotebookLM simply dumped the finished WAV file of the conversation on your desktop, and you then had to work out how to share it with people. That wasn't too hard, but it required a few extra steps and other services. Now NotebookLM will give you a publicly shareable URL for every file you create, which makes life so much easier. NotebookLM is one of the most impressive AI tools we've used yet and everyone who's tried it out in the office has had a similar response: It doesn't sound like AI, and it's almost too real. Previously you could only upload documents like PDFs or text files, but now the ability to just drag and drop any YouTube video on the internet adds a whole new dimension to NotebookLM. There are so many possibilities with an AI tool that can quickly scan information and create an easy-to-follow discussion. Not only can you now quickly create AI-generated conversations that can pass as professional podcasts, but it also opens the door for a much improved educational experience. In fact, Google say it's already being used by students and lecturers at Boise State University to assist learning. We're only just at the beginning of NotebookLM's development and the future of this AI audio tool is already looking incredibly bright. It's not just Google working magic with AI audio, however. Yesterday at Meta Connect 2024, Meta unveiled new audio translation tools that will automatically dub and lip-sync Instagram and Facebook Reels. With all of these impressive AI audio tools emerging, it's going to be an exciting couple of years as we see the true potential of this technology.

Google's NotebookLM can now turn audio recordings and YouTube videos into interactive podcasts

There's an Audio Overview of this very article embedded at the bottom. In all honesty, it does accurately convey what's written here, though the facts are sandwiched between a lot of fluff that feels like an ad for NotebookLM -- made by NotebookLM. You might find the written version much more concise. Key Takeaways Google's Gemini-powered NotebookLM promotes healthy usage of AI for learning with key features that help surface key insights, breakdown complex material, and more. The tool's Audio Overviews feature can turn a block of text into an interactive spoken-dialogue discussion, making it a useful tool for those that learn the best when they listen to information rather than reading it. The feature can now import information from audio files and even YouTube videos, complete with an option to share your Audio Overview with others. OpenAI's ChatGPT changed the homework game when it first blew up in January 2023. The AI assistant, which was abused for its supposed all-knowing capabilities, served more like a tool for academic dishonesty than a tool to help students learn. In this instance, don't hate the game, hate the player. No judging though, you gotta do what you gotta do. The potential to abuse subsequently-released AI assistants and chatbots in academic settings is still prevalent, though Google is, to some extent, promoting healthier usage, with tools to help students acquire knowledge instead of plagiarism strikes. The tech giant's NotebookLM is a perfect example. Related Google Gemini in NotebookLM is finally ready to turn your textbooks into interactive podcasts Listen your way to learning The Gemini-powered tool, which allows users to upload documents or plain text and have the AI tool convert the material into digestable and simple-to-understand excerpts, can do a lot more. It can help surface key insights, help understand complex answers, and even turn your homework into an interactive podcast with Audio Overviews. The latter, which is a huge boon for those that learn the best when they listen to information rather than reading it, is now getting a solid boost. The tool, which can generate a spoken dialogue discussion between two people from the source material, could previously import information from user-uploaded PDFs, text documents, documents from Google Docs and Google Slides, pasted-in text, and links to websites. Via links, the tool could only import visible texts, though that seems to be changing now. According to a new blog post by the tech giant, users should now be able to YouTube video URLs directly into the tool, allowing them to "easily summarize key points and arguments from educational YouTube content." While not mentioned, it is likely that the tool uses said YouTube video's captions to analyze and summarize it for you, similar to what Gemini's "Ask about this video" tool does. You can now share your Audio Overviews Elsewhere, users should now also be able to upload audio files in coding formats like .mp3 and .wav. Further, while users could previously download the Audio Overview file for offline listening or to share it with others, the tool will now natively allow users to share the audio file with other users. "You can now click on the share icon on the Audio Overview generated in NotebookLM and it will generate a public URL that you can copy to share with others," suggests the tech giant. The sharing option isn't live for us yet, though it will likely show up near the Overview's progress bar, as seen in the highlighted section in the image above. And now, for those looking for a more auditory experience, here's an Audio Overview of this article: