
How to create an AI Podcast with NotebookLM?
With NotebookLM, you can create an AI Podcast from your research, in this article let's see how.

Kaushal Joshi
Oct 09, 2024 • 6 min read
NotebookLM is a powerful AI-driven research and note-taking tool developed by Google. Its standout feature is NotebookLM Audio Generation, which allows users to turn their research into engaging podcasts. In this blog, we’ll go through creating a NotebookLM podcast on technical topics, using React Server Components as an example.
What is NotebookLM?
NotebookLM is a research and note-taking tool developed by Google. Unlike other AI tools, NotebookLM is designed to work specifically with user-provided data, personalizing outputs based on the documents you upload.
It takes various sources in a wide variety of formats and trains itself specifically on that data. This helps you get answers specific to the sources you provide, increasing your productivity significantly.
I have written another blog that explains how to use NotebookLM in brief. It talks about NotebookLM’s features, use cases, starting guide, etc. Today we’ll focus on how to create a podcast with AI using its audio generation feature.
What is Audio Generation in NotebookLM?
The NotebookLM Audio Generation feature is an innovative tool that enhances how you consume and interact with your research materials. It takes various source documents — research papers, blog posts, or even videos— and transforms them into an interactive podcast.

Instead of reading through lengthy texts, you can listen to a dynamic conversation between two AI-generated voices discussing the content, summarizing, and analyzing the key points.
How to Create a Podcast with NotebookLM?
Now let’s get to the interesting part of creating a NotebookLM podcast. We will create a podcast episode on the React Server Component. It’s totally fine if you aren’t aware of this at all. It’s just a topic for the podcast, and won’t affect the podcast creation process at all!
Step 1: Get started with NotebookLM
Getting started with NotebookLM is as simple as signing in with your Google account.
Sign in to NotebookLM using your Google account.
Click on
Create New Notebook
and organize your source material in this notebook. Consider a notebook as a workspace that includes everything around one topic.Rename the notebook for clarity by clicking on
Untitled Notebook
in the navbar.
Step 2: Add your sources
The quality of your NotebookLM podcast depends on the sources you upload. Include reliable, high-quality resources like documentation, blog posts, research papers, and YouTube videos. You can upload up to 50 documents, each with a 50,000 character limit.
NotesbookLM supports various formats as sources, including PDFs, markdown or simple text files, web links, YouTube video links, Google Docs, Google Sheets, etc. You can upload up to 50 documents per notebook, where every document must not have more than 50,000 characters.

In our case, let’s add some in-depth videos and blogs published around React Server Components. Also, we’d add links to some GitHub repositories, GitHub discussions, etc.
Step 3: Review and Select Key Sources
Once your documents are uploaded, NotebookLM will summarize them for you. Review the summarized content to ensure the AI focuses on the most important topics related to React Server Components.
If certain documents seem irrelevant or off-topic, you can choose to exclude them from the processing step. This will fine-tune the final podcast with AI, keeping it focused on the most valuable insights.

We have only added a handful of resources, and all of them are important, hence we won’t exclude anything.
Step 4: Generate the Audio Overview
When you're satisfied with your selection of sources, look for the Generate Audio Overview
button. This will prompt NotebookLM to create a podcast-style conversation between two AI voices, discussing React Server Components.

Depending on the complexity of the sources you’ve uploaded (length of articles, number of documents), it may take a few minutes for NotebookLM to generate the audio overview. Larger documents will take longer to process.
Download and Share the Podcast
Once the AI finishes generating the audio, NotebookLM will provide you with a downloadable file. This file will contain the entire discussion on React Server Components, ready to be edited or published.

You can also share the podcast publicly with others. Click the share button beside the podcast title. Grant the access to share it publically. Now it will generate a public link. Click Copy share link
to copy the link to your clipboard. The listener must need to sign in with their Google account to listen to the podcast.
How to Get the Most from NotebookLM Podcasts?
Here are some tips to maximize the podcast quality by NotebookLM:
Curate high-quality sources: To ensure an insightful discussion, use trusted sources like official documentation, well-researched blog posts, and expert articles.
Experiment with Source Diversity: Mix technical articles with opinion pieces or tutorials. This can result in a more dynamic, engaging conversation in the podcast.
Personalize with Manual Input: If you want more control, you can add personal notes or questions to your sources. For example, if you want the AI to explain a specific challenge related to React Server Components, make sure that note is included in the documents uploaded.
Copyright and Plagiarism Considerations for Podcast Creation with NotebookLM
In my opinion, creating a podcast with AI using NotebookLM is useful to help you get better insights and learnings from the sources you provide. It’s helpful when you learn better by listening than reading. So you can listen to it while you’re completing your chores, traveling, working out, etc.
Nonetheless, it’s very important to be aware of copyright and plagiarism concerns. Here’s a simple breakdown of what you need to keep in mind in case you plan to share a podcast generated by NotebookLM publically:
Respect Copyright Laws: When uploading source materials into NotebookLM, ensure you have the legal right to use them. Many documents like articles and books may be copyrighted, so always check the terms of use or licenses. Using unauthorized content without permission could violate copyright laws.
Avoid Plagiarism: Always attribute the original authors when generating a podcast based on external sources. It's essential to credit ideas, text, or research from external content to avoid plagiarism. Additionally, ensure the AI doesn’t replicate large portions of copyrighted material verbatim.
AI-Generated Content: While AI can help generate content, you are still responsible for ensuring that the podcast complies with copyright laws. Always review the audio before publishing to avoid using unauthorized excerpts. Also, add your unique insights to make the podcast original and engaging.
If you're uncertain whether certain materials can be used, consider contacting the copyright holders for permission. This is particularly important for any proprietary content or content behind paywalls.
Wrapping up
Creating a NotebookLM podcast is an innovative way to turn your research into audio content. Whether you want to discuss React Server Components or any other topic, NotebookLM’s audio generation feature can simplify the process, helping you create and share valuable insights efficiently.
Are you using NotebookLM in your daily life? How are you using its Audio Generation feature to create Podcasts with AI? I’d love to know. I am most active on Peerlist and Twitter if you want to say hi!
Be sure to check out Peerlist if you're searching for a new job or have an interesting side project you'd want to share with the world.