Use AI to Build the Business and the Life, You Actually Want. Practical insights on AI, identity, and growth for entrepreneurs who are done playing small. One email a week. No noise.

Home › Playbooks › How to Create a Podcast with NotebookLM

How to Create a Podcast with NotebookLM

This playbook will show you how to turn your notes and sources into an audio-learning experience using NotebookLM, which can be particularly useful for students and researchers dealing with a lot of information.

Step 1: Access NotebookLM

  • Head to NotebookLM in your browser.
  • Sign in with your Google account.
  • If you’ve used it before, your existing notebooks will appear. If it’s your first time, you’ll be prompted to create a new notebook.

Step 2: Create a New Notebook

  • If prompted, click “Create” to start a new notebook.

Step 3: Add Your Sources

  • Immediately after creating a notebook, you’ll see the “Add Sources” page.
  • Upload your sources: You can upload up to 50 sources, including links, Google Drive files, MP3s, and more.
    • Note: The podcast and any responses will be based solely on the sources you provide, and your data won’t be used for AI training.
    • Tip for YouTube videos: If you’re using a YouTube video link as a source, be aware that if the video doesn’t have a transcript, the source link might turn red, indicating it can’t be used.
  • Click “Insert” once you’ve selected your sources.
  • Add more sources by clicking the ‘+’ button on the left side of the interface. The more sources you add, the better your notes, summary, and ultimately, your podcast will be.

Step 4: Familiarise Yourself with the Notebook Guide

  • Once your sources are in, the Notebook Guide should pop up, providing a one-paragraph summary of all your sources. If it doesn’t show up right away, you can find it in the bottom right corner of your notebook.
  • The Notebook Guide has several sections to help you:
    • “Help Me Create” section: Located above the summary paragraph, this AI tool offers options to create an FAQ list, a study guide, a table of contents, and other useful summary breakdowns.
    • “View Chat” option: Below the summary, this allows you to view any past conversations you’ve had about your sources.
    • Suggested Questions section: To the right, NotebookLM generates questions based on your provided sources.

Step 5: Navigate to the Audio Overview Section

  • Once you’re satisfied with the number of sources you’ve added, head over to the “Audio Overview” section. This is the dedicated area where NotebookLM will begin converting your sources into a podcast-style conversation.

Step 6: Generate Your Podcast

  • In the Audio Overview section, click “Generate”.
  • Warning: Generating a podcast does take a few minutes, so you can switch to another tab while it runs in the background.

Step 7: Listen to Your Personalised Podcast

  • Once it’s ready, your personalised podcast will play. It’s designed to sound like a conversation between a male and female host, complete with pauses, ‘ums,’ and even jokes.
  • The length of the podcast can’t be adjusted, but most tested examples ranged from 5 to 15 minutes. For instance, a podcast generated from just two sources impressed the creator by stretching over 8 minutes, significantly longer than an original 3-minute video.

Step 8: Manage Your Generated Podcast

  • If you’re happy with the audio, you have several options:
    • Download the audio.
    • Change the speed.
    • Share it as a link.
    • Embed it on your website.
  • Remember to name your notebook to keep things organised, and you can return to it anytime.

Bonus Tips and Considerations for Audio Overviews:

  • AI-Generated Content: Both the Audio Overviews and the voices are AI-generated, so there might be inaccuracies or audio glitches.
  • Repetitive Phrases: The podcast hosts often say “deep dive” multiple times, which can become distracting or confusing if not relevant to the context of your sources.
  • Language and Speakers: As of the video’s creation, Audio Overviews are only available in English. You also cannot control the number or gender of the speakers.

Share this post: