Meet Google’s automatic podcast generator

What Does NotebookLM Actually Offer?

NotebookLM functions as an instant podcast generator where two virtual hosts discuss the content you upload, whether it’s an article, book, presentation, or even a transcription of a phone call. Using advanced speech recognition and machine learning technologies, Google has managed to create an almost realistic experience where the virtual hosts sound like humans delving into and processing the content into a rich conversation.

The new “Audio Overview” feature offers much more than meets the eye: the machine not only reads the text but actively manages a conversation, analyzing the material and presenting it in an innovative and entertaining manner. It allows users to digest and understand information in lengthy documents or videos through AI-generated audio conversations.

Personal Control Over the Final Product

Until now, audio reviews automatically generated AI conversations from the sources you provided. However, since conversations sometimes revolve around unimportant content, Google is now introducing an update that enables users to focus the “hosts” on specific topics within the content, delve into certain issues, or skip others. Thus, users can take on the role of producers and guide the discussion according to their preferences. Customizing audio summaries may also help reduce hallucinations to some extent, meaning those instances when AI creates content on its own rather than relying on solid data.

In addition to the customization option, users receive background listening during audio reviews. This allows them to continue working within NotebookLM, querying their sources and exploring relevant quotes while the audio plays in the background.

Dive Deep Without Wasting Time Reading

The standout advantage of NotebookLM is its ability to save valuable time. Instead of sitting down to read a lengthy document, users can upload the content to the system and receive a rich audio “summary” that includes all the key points within minutes, designed for listening on the go, during workouts, or while engaging in other activities.

In an age where information floods in from all directions, this capability can save countless hours of reading and deliver the essence of the content—only what is truly essential. However, it may be more useful for specific purposes like studying and memorizing material for exams, rather than for leisure enrichment.

While initial interest in the product was primarily for educational and research purposes, the new features added by Google recently began attracting businesses and organizations to try it out. Google now claims that over 80,000 organizations are using NotebookLM and views it as an opportunity to explore monetization.

In hopes of harnessing this power, the company launched the NotebookLM Business pilot program on Thursday. Under the business pilot, Google’s team trains organizations interested in using NotebookLM on how other businesses are leveraging it. Companies can apply for the pilot, and if accepted, they will receive early access to product features, training, and email support.

Not Everything is Rosy

For many, this represents a gateway to quick and deep learning—but like any technology, it comes with unique challenges. When it comes to traditional podcasts, a significant part of the experience stems from the chemistry between the hosts, the personalities they bring to the conversation, and their spontaneous reactions. In the NotebookLM platform, this element is absent—the virtual hosts may claim to manage a compelling conversation, but the human experience that podcast enthusiasts cherish remains largely outside the equation.

Additionally, there are still instances of inaccuracies in content, as AI relies on what it is given and cannot delve deeper than what has been fed into the system. This challenge is currently at the heart of AI development, as developers strive to ensure high levels of accuracy and flawless quality, even with information that exists on the fringes of the text.

The platform is being gradually rolled out and is still in a trial phase, which means support for languages other than English is still shaky. The interface does support Hebrew, and it’s even possible to upload files in Hebrew; however, the outputs themselves are currently only available in English.

Shortly after its launch last month, many began sharing audio summaries of their content on social media. Data from the traffic analysis platform SimilarWeb shows that NotebookLM saw a rise of over 371% in its traffic in September, reaching 3.07 million monthly visits compared to 652,000 the previous month. Given this impressive growth, we expect it’s only a matter of time before support for additional languages, including Hebrew, is introduced.

A Look to the Future: Will NotebookLM Change the Podcasting World?

As technology continues to evolve, one can envision NotebookLM or similar tools interfacing with live databases and performing in-depth analysis on information from multiple sources. In the future, the platform may also include tools that enrich the conversation with additional dynamic interactions, and perhaps even incorporate video clips and graphics that accompany the audio.

If the future allows users to connect with knowledge more quickly and accurately without sacrificing depth and relevance, one can certainly see how tools like NotebookLM might change the way we consume content—and possibly expand the boundaries of podcasting itself. The big question is whether listeners will agree to forgo the personal human touch and settle for dry information alone. We’re not so sure about that.