Initially introduced as Project Tailwind at Google I/O 2023, NotebookLM, an experimental tool developed by Google Labs, is now accessible in the U.S. for users aged 18 and above. The tool utilizes Gemini Pro, the tech giant"s AI model designed to facilitate document understanding and reasoning, broadening its utility across various tasks.
The latest release of NotebookLM introduces over a dozen new features aiming to assist users in enhancing their thought processes and productivity. Notably, it includes a noteboard space for saving notable content from conversations, sources, or personal notes.
Additionally, it offers suggested actions based on user activity, such as summarizing text, deciphering technical language, or suggesting related ideas. While composing a note, it offers tools to elevate your writing style and proposes correlated concepts sourced from your ongoing notes, ensuring coherence and relevance to your written content.
Another feature enables users to organize their curated notes into structured documents by selecting notes and instructing NotebookLM to convert them into new formats such as thematic outlines or study guides.
Below is the detail of all the new features:
- Increased source limit: Notebooks can now include up to 20 sources.
- Increased word count: Sources can now include up to 200,000 words.
- Write individual notes: Adding a note will now create an independent new note instead of adding to a single notepad.
- Noteboard: You can now see all your written and saved notes pinned to the Noteboard space above the chat box.
- Save chat responses as notes. NotebookLM responses can be pinned to the noteboard for your reference later.
- Citations saved to notes. When chat responses are saved as notes, it retains the original citations.
- Jump to citation in source. Clicking on the citation number in a chat response or a saved note immediately takes you to the original quote in the source.
- Expand/collapse source section. You can now hide the source if you want to concentrate exclusively on note-taking.
- Focus the AI on selected sources. You can chat with a specific set of sources in your notebook by selecting them individually in the source sidebar.
- Source guides integrated into individual sources. Source guides, including summaries and key topics, now appear at the top of each source in the source sidebar.
- Shared notebooks: Notebooks can be shared in both viewer and editor modes. Viewer simply shares your assembled sources; editor lets you collaborate on creating notes.
- Suggested questions: NotebookLM now automatically suggests followup questions for you, based on your recent conversation history and the content of your sources.
- PDF support: PDFs can now be uploaded as sources in NotebookLM, up to 100MB or the word limit.
- Copied text support: You can now copy and paste text to create a new source and add/edit the title upon creation.
- Example notebooks: Learn more about how to use Notebook before uploading your own sources. See example notes and try Q&A questions.
- [Coming next week] Suggested actions. NotebookLM gives you a palette of preselected actions that can be used to transform selected text or notes. (Currently only supporting selected notes.) Initial suggested actions include:
- Combine to a single note. Gather all your notes into a single unified note with one click.
- Critique. Ask NotebookLM to give you constructive feedback on your prose or argument.
- Summarize. Create a concise, easy-to-read overview of the content of multiple notes.
- Create outline. Convert your selected notes into an outline, organized around topics.
- Create study guide. Create an on-the-fly study guide based on your notes, including key questions and a glossary.
- Suggest related ideas. NotebookLM can prompt you with related ideas from your sources based on the content of the selected notes.
NotebookLM operates as an AI-native application, leveraging modern technology capabilities. As an ongoing experiment, it continually improves based on user feedback. Notably, user data is not utilized to train NotebookLM, ensuring the privacy of any sensitive information within sources.
To try NotebookLM, you can head over to the webpage here.