Google announced on Thursday that it is extending the availability of NotebookLM, its AI-powered note-taking assistant, to over 200 new countries, approximately six months after initially launching it in the United States. NotebookLM, driven by Google’s advanced multimodal LLM Gemini 1.5 Pro, has been enhanced with new features and additional languages, making it easier for users to leverage AI for generating summaries and posing questions based on their documents.
Among the countries gaining access to NotebookLM are Australia, Brazil, Canada, India, and the UK, in addition to 208 other countries and territories. The AI-assisted app’s interface now supports 108 languages, such as Arabic, Assamese, Bengali, Cantonese, Chinese, Dutch, French, German, Hindi, and Hinglish. Furthermore, it facilitates sources and chat in 38 languages, including Arabic, Bengali, Chinese (simplified and traditional), Dutch, French, German, Hindi, Japanese, and Spanish.
Initially introduced as Project Tailwind at Google I/O 2023, NotebookLM became available to select users in June last year. The AI-driven tool assists in generating summaries and responding to questions from documents, transcripts, notes, and other materials uploaded by users. Unlike traditional AI chatbots like ChatGPT, which often deviate from provided sources, NotebookLM adheres closely to user-uploaded content.
Google has expanded NotebookLM’s capabilities to include sourcing content from Google Slides and web URLs, complementing its existing support for Google Docs, PDFs, and text files. This enhancement allows users to create notes or inquire about content within their documents, be it imagery or text, as well as explore online materials.
Early U.S. users anticipated support for traditional note-taking applications such as Evernote and Google Keep. However, Raiza Martin, Senior Product Manager for AI at Google Labs, indicated in a recent virtual roundtable that Google prioritizes establishing the product’s core value before pursuing broader integrations.
“Down the road, you’ll hopefully see these types of integrations,” she stated.
Google has also introduced inline citations, enabling users to view supporting passages within their sources, verify AI-generated responses, and read original texts for additional context. Previously, citations were positioned below the assistant-generated responses.
Another feature, Notebook Guide, assists in transforming content into various formats, including FAQs, briefing documents, or study guides.
Steven Johnson, Editorial Director at Google Labs, mentioned that NotebookLM was designed with input from authors, students, and educators. The tool’s source-grounding architecture has been adopted by early users to enhance their research and writing workflows.
Google highlighted that NotebookLM has been utilized for creating hyperlocal newsletters, summarizing interview transcripts, developing grant proposals, and managing descriptions in fantasy worlds.
Martin assured that Google does not use user-uploaded data on NotebookLM to train its algorithms.
“We get this question a lot because users want to be able to use it with work or school documents,” she noted. “Your data does stay private to you.”
During its Google I/O 2024 keynote in May, Google showcased an early prototype of Audio Overviews for NotebookLM. This feature utilizes the Gemini model to scan uploaded materials and generate podcast-style discussions. Gemini 1.5 Pro allows NotebookLM to accommodate up to 50 sources per notebook, with a word limit of 500,000 per source.
The global expansion of NotebookLM is poised to compete with numerous platforms that currently provide AI tools for summarizing documents and answering queries. While most of these platforms require payment for their services, Google offers NotebookLM free of charge, leveraging its extensive resources.
