Focus: Deep understanding and interaction with your own documents. It’s like having an AI assistant dedicated to helping you make sense of your research, notes, and other materials.
Key Features:
Upload and analyze various document formats (PDFs, Google Docs, etc.)
AI-powered summarization and information extraction
Interactive question answering with source citations
Content generation based on your sources (summaries, timelines, etc.)
Audio overviews for on-the-go information consumption
Strengths:
Excellent for research, learning, and writing tasks that require you to work closely with specific documents.
Strong emphasis on accuracy and source grounding, reducing the risk of AI “hallucinations.”
Designed to help you organize your knowledge and gain deeper insights from your information.
Focus: A versatile AI model that can perform a wide range of tasks, including text generation, code generation, image creation, and more. It’s designed to be a general-purpose AI assistant.
Key Features:
Multimodal capabilities (text, code, images)
Advanced reasoning and problem-solving abilities
Can be integrated into various applications and workflows.
Strengths:
Highly adaptable and can be used for a broad range of tasks.
Excels at creative content generation and complex problem-solving.
Can be used to power various AI-driven features in Google products.
NotebookLM is like a specialized research assistant who meticulously analyzes your documents and helps you extract the key insights.
Gemini is like a versatile general assistant who can help you with a wide array of tasks, from writing emails to brainstorming ideas.
Use NotebookLM when you need to deeply understand and work with your own documents.
Use Gemini when you need a more general-purpose AI assistant for various tasks, including creative content generation and problem-solving.
It’s worth noting that Google is constantly evolving both NotebookLM and Gemini, so their capabilities may expand and overlap in the future.
When you upload a new source to NotebookLM, the app instantly creates a source overview that summarizes the document and offers key topics and questions to ask.
Think of it as a research assistant that helps you better understand the source material.
Right now, NotebookLM supports the following source types:
Google Docs
Google Slides
PDF, Text and Markdown files
Web URLs
Copy-pasted text
Youtube URLs of public videos
Audio files
Each source can contain up to 500,000 words, or up to 200MB for uploaded files.
Avoid uploading documents you don’t have the applicable rights to.
NotebookLM can’t delete or edit your original files in Drive. When you import Google Doc or Google Slides, the app makes a copy of the original file. NotebookLM doesn’t keep track of changes to the original doc automatically so you have to manually re-sync the imported Google Docs and Slides in the source viewer.
You will only see the ”Click to sync with Google Drive” button in the left hand side source viewer if the original file has been updated since the last time you viewed the current source.
Limitations:
The ‘Click to sync with Drive’ button in NotebookLM will only be displayed if you have write access to the original Google Drive file
Other types of sources will need to be manually deleted and re-uploaded, as NotebookLM only keeps a static copy of the file at upload time
Only the text content of the given webpage will be scraped for use as a source; images, embedded videos or nested webpages will not be imported
Paywalled webpages or those that have disabled web scraping are not supported
Only public YouTube videos with captions (either user-uploaded or auto-generated) are supported
Only the text transcript of the video will be imported as a source
Recently uploaded videos (<72 hours) may not be available to import; you may try again later
Videos with no speech are not supported
Sources will be auto-deleted from your notebook within 30 days if the video itself is deleted or made private
No limit for length of video, unless the caption file contain over 500,000 words
Question:
My import still failed, why?
This is possible for a number of reasons, but the most common are: the content isn’t a valid YouTube link, the video is potentially unsafe, the content does not have a captions file that I can read, or the video language is currently not supported.
The audio file will be transcribed during import and its text will be saved to use as a new source
Supported audio file types include MP3, WAV, among others
Audios with no speech are not supported.
Nearly all languages are supported for audio import, however import may still fail if the audio is low quality.