What you need to know
- Google’s NotebookLM condenses large amounts of documents and notes into 60-second TikTok-style vertical videos.
- This feature is powered by Gemini 3.1 Flash-Lite Image, Google’s fastest and most cost-effective image generation model with a 4-second render time.
- Google is also previewing a robust full video generation model that supports natural language editing, multimodal input, and native audio sync.
No one likes wading through a huge stack of notes or a 50-page research document when time is limited. Google knows this, which is why it’s rolling out a major upgrade to NotebookLM that turns uploaded documents into AI-generated 60-second bite-sized videos.
For the uninitiated, NotebookLM started out as a simple AI research assistant, but is now a powerful learning tool. Our newest feature, Short Video Overview, is an expansion of the Cinematic Video Overview that we previously discussed. But the real star of the show is the engine that powers it, Google’s new Nano Banana 2 Lite.
Nano Banana 2 Lite, officially called Gemini 3.1 Flash-Lite Image, is Google’s fastest and most cost-effective image generation model to date. The image generation time was a whopping 4 seconds, and the visual quality was significantly improved compared to the original Gemini 2.5 Flash image (the first Nano Banana). This engine combines world knowledge with a solid, consistent character. Quickly sketch accurate scenes, develop storyboarding tools, and render legible typography for localized ads without compromising object fidelity.
Short video summaries in NotebookLM enable this model, condensing sources into vertical videos packed with narrative explanations and educational animations. It’s designed for people revising notes, presentations, or long PDFs who only want to know the important points. Before generating, simply select “Short” from the output formats (along with Explainer and Cinematic), select a specific source, and set the focus of the video using custom prompts or suggested topics.
In addition to this, Google has pushed Gemini Omni Flash into public preview. This model is intended for complete video generation with conversational editing (scene relighting and natural language text exchange) and multimodal input. Each video output has native audio pairing to synchronize on-screen text and dynamic movement.
Short video summaries are not yet available to everyone. This feature will begin rolling out to English-speaking users 18 and older in the coming weeks. Once published, it will be accessible from the web, Android, and iOS, whether you’re logged in to your regular consumer account or Google Workspace.
Android Central’s opinion
Whether you’re a student cramming for a final exam or just hate reading thick papers, TikTok-style summaries can be a great way to turn a 50-page PDF into an easy-to-understand content. But are we really so allergic to actually reading that we need an AI to hallucinate 60-second clips of our notes? Google wants to pitch the service as the ultimate productivity hack, but it feels ominously like a crutch to maintain our rapidly shrinking attention spans.
