This Google AI Tool Can Turn Your Research Into a ‘Podcast’

When it comes to Google’s AI tools, the first thing you might think about is the Gemini assistant app or some cool tricks you can do in Google Photos . However, Google has another powerful AI tool that is great for collecting and collating notes on any topic. It’s called NotebookLM .

The purpose of NotebookLM (LM, of course, stands for Language Model) is to help you make sense of what you’re researching. It will help you by creating summaries, answering questions about the documents you’ve collected, and connecting data points together where needed. As is typical with this type of AI, it is still considered an experimental feature and you will need to double-check anything it produces for hallucinations .

Still, it’s a potentially useful tool, and it has another new feature: audio reviews . Essentially, this feature allows you to create short podcasts using AI podcast hosts that summarize all the information you’ve collected. Whether you’re researching 19th-century literature or which new phone to buy, Audio Review will present it in a friendly and accessible way, and the results can be quite impressive.

You can access NotebookLM and the underlying Gemini models for free using your Google account. Try this with your own research notes and be prepared to be surprised by how natural the resulting clips sound.

Getting started with NotebookLM

Adding sources to NotebookLM. 1 credit

If you’re new to NotebookLM, you can sign in to your Google account to get started. You’ll see some example projects (called notebooks) that you can play with. Open, for example, “The Invention of the Light Bulb ” and try asking the AI ​​questions about the work of Joseph Swan, Thomas Edison and others.

The Q&A routine will be familiar if you’ve used something like ChatGPT, Gemini or Copilot, but here the training data used for the model is only the sources you’ve loaded into your notebook, not the wider net. However, you can point NotebookLM to web links, and also add plain text, PDFs, Google Docs, and Google Slides (each notebook can have up to 50 sources).

To start building your own notebook, click the NotebookLM header (top left), then select New Notebook . You will be prompted to start adding sources to a new notebook; you can upload documents from your computer, select them from Google Drive, point NotebookLM to web links, and paste text from the clipboard. You can add additional sources later by clicking the + (plus) button next to “Sources” in the left pane.

Once you start adding sources, NotebookLM will begin to summarize them and then suggest questions you might want to ask. You can click on individual sources to view them individually, and follow the on-screen links to create an FAQ, study guide, table of contents, timeline, or reference document for everything you’ve collected.

Creating an Audio Review

Your podcast will appear in the notebook overview. 1 credit

For the purposes of this guide, I have downloaded a PDF of recent research on how smartphone apps can be developed to measure blood pressure without using a cuff. I’ve also included the accompanying press release as a web link to give NotebookLM a little more material to work with.

Right off the bat, the AI-proposed question suggests how the app uses existing smartphone technology, what the potential benefits might be, and what the challenges of developing this technology are—and overall, NotebookLM provided answers that made sense and were accurate. based on the material provided. (When you use the app, the answers you receive are accompanied by citations that take you back to the sources of that information.)

To create a podcast (or audio review), you’ll need to open the notepad guide (the link will be on the right if you’re on the chat screen) and then click “Create” on the right. There are currently no settings for you to play with – you get the same two podcast hosts, one male voice and one female voice, speaking English every time. After a few minutes of thinking, the sound will start playing and you can also download the file if necessary.

Impressive, but not without reservations

In my test, Audio Review did an excellent job of summarizing the research on a blood pressure app with a high level of accuracy, creating what sounds like a natural podcast conversation between two people, although it missed some nuance and was a bit generic. in some places. As a technical demo and summary tool it’s really impressive, but if details really matter in your work, I wouldn’t rely entirely on AI at this stage.

More…

Leave a Reply