I have been meaning to try out Google Notebook LM for some time. It seemed like an intriguing idea when I first read about it, but at launch, it was only available in the US, and I was too busy to play with the application when it launched in the UK.
That changed a couple of weeks ago when I found out about the hype surrounding new functionality added to the application, which allowed it to generate a two-person conversation podcast generated by AI. I had to try it.
To cut to the chase, I was impressed, and I think Google may well be developing a killer large language model application with Notebook LM. I will explore this in another blog post.
Today, I will take you through what Google Notebook LM is and why its ability to generate an audio podcast is such a breakthrough before finally explaining the process of creating a podcast.
What is Google Notebook LM?
Google Notebook LM is a note-taking application similar to Google Keep or Microsoft OneNote. One key difference is the inclusion of a large language model, which can summarise your notes, generate FAQs, or generate your questions through the application’s chat functionality.
There are note-taking applications that have implemented large language models inside them, including Notion, Mem, and even Obsidian with a community add-on. These applications have an advantage over Google Notebook LM in that they can be used to create a Personal Knowledge Management system.
Another limitation of Google Notebook LM is that it currently limits you to 50 notes in one notebook.
For me, the question is, does the introduction of the podcast generator overcome these limitations?
Why is Google Notebook LM such a big deal?
Firstly, it’s kind of cool and a bit terrifying that a large language model can take the information contained within a notebook and create a podcast with AI-generated voices that make sense.
Secondly, I can see a use case. Let’s imagine that you are at school studying Algebra. You upload your notes and some useful web pages you find on algebra on the Internet to the algebra notebook you have created in Google Notebook LM.
You can read the summary and FAQs and even have a conversation with the content you added to the notebook, which is all very useful in itself.
However, the podcast you generated from the notes within your notebook gives you a different perspective, which might help you understand how algebra works.
How to create a podcast
In this section I will take you through the process of creating a podcast in Google Notebook LM. I was originally planning to turn one of my earlier posts into a podcast, but Now I think it would be more interesting to use the previous algebra example. I have no course book, but we should still be able to recreate the Internet resources of the Algebra example I mentioned earlier.
If you haven’t tried NotebookLM yet, visit https://notebooklm.google/ and click the try NotebookLM button. You will need to log in with your Google profile.
Once you have an account, open the NotebookLM web app in your browser and create a new notebook. I then searched with the query “learning algebra.” Add links to websites or YouTube videos that would help you learn algebra.
I tried the following two resources. https://thirdspacelearning.com/gcse-maths/algebra/maths-algebra/ and https://www.youtube.com/watch?v=PVoTRu3p6ug
Once you have your resources, click on the Notebook guide. You can click on Customise to modify the generated podcast. As this example is for children learning algebra, I told it to aim at beginners learning Algebra. Once I was happy with the change, I clicked the generate button.
I have embedded the generated podcast from my notes below. The original was created as a WAV file, and I converted it to MP3 before uploading it to my webserver. No other changes have been made to the file.
Conclusion
You should now have a better understanding of Google Notebook LM and why I think the ability to create your own AI-generated podcast based on your notes is such a breakthrough.
For further reading I would suggest our introductory guide to what is a Large Language Model.
Enjoying our content, why don’t you join our monthly Newsletter? We will update you on the latest content we have published and what currently intrigues us in the world of technology, computing and gaming.