Meta Challenges Google with Open-Source AI Podcast Generator NotebookLlama

Meta takes on Google's NotebookLM with new NotebookLlama -- and its totally free

Well that didn't take long. Mere months after Google Labs launched its now viral NotebookLM product, king of the open source world Meta has announced a free customizable AI podcast maker. NotebookLlama features the same kind of functionality as the Google version, although it's rougher around the edges and definitely not as pretty. The key principle of this type of product, and what makes them so popular, is the ability to take raw input like a PDF, doc file, or web URL and convert that to an engaging podcast episode featuring two audio hosts chatting about the subject matter. It's super clever, incredibly good at distilling complex topics and a fun way to do learning or research. NotebookLlama, while not as polished as Google's offering, has the benefit of being open source and more adaptable, making it a viable alternative. The new NotebookLlama from Meta is nowhere near as polished in terms of features or results, but it's designed as a DIY manual, aka recipe, for anyone with the smarts who wants to create their own NotebookLM-style AI podcast product. And that's a big deal. The whole package is of course built around Meta's Llama family of language models, which can be accessed from various sources across the Web. What's interesting is the architecture doesn't rely only on huge memory-hogging models, but has also been designed to work with the smallest models in the Llama range. That means it should also be usable on modest-sized computer equipment. Another big deal. The initial audio demo is very impressive, especially considering the product is only a few hours old. The 'recipe' uses off-the-shelf technology to power the voices, and it's surprisingly good at conveying the same kind of emotive vocal nuance that makes the Google product so compelling. There are even some expressive ums and ahs in there, which helps to balance out the slightly dodgy audio quality, where the volume sinks or word spacing stutters. The product page on Github links to some additional resources for people to try out better voice tech options. Hopefully, this means we'll see an improvement in the quality and range of vocals in the near future, once the open-source community gets to work. The development team is also asking for dev help to add more media sources for the podcast engine to process. At the moment it's only PDF files, but the team seems to be hopeful of eventually adding web links, audio and even YouTube links to match Google's original. It's shocking not just how fast this new product has been released, but how good it is for a first iteration. The Meta offering is not the first clone to hit the market, Gabriel Chua released Open NotebookLM on a Hugging Face Space a few weeks back, again based on a Llama model. But that is nowhere near as slick as this new version, which means NotebookLlama is sure to grab the imagination and support of developers everywhere as the potential starts to sink in. Get ready for a flood of AI podcast makers to hit the scene soon.

NotebookLlama: Meta has its own open-source NotebookLM

Meta has introduced an open implementation of the generate-a-podcast feature that Google offers in its NotebookLM platform. Named NotebookLlama, this new project utilizes Meta's own Llama models for most of its processing. Similar to NotebookLM, NotebookLlama allows users to create podcast-style digests from text files, such as PDFs of articles or blog posts. NotebookLlama starts by creating a transcript from a given file -- for example, a PDF. The system then adds elements like dramatization and interruptions to make the generated content feel more like a conversation. After that, it uses open text-to-speech models to convert the transcript into audio. The current output quality of NotebookLlama's generated podcasts is still rough compared to Google's NotebookLM. The voices have a noticeable robotic quality, and they often talk over one another at odd times. However, Meta's researchers point out that improving this quality is possible with stronger text-to-speech models. On NotebookLlama's GitHub page, they note, "The text-to-speech model is the limitation of how natural this will sound." One possible improvement for the project, according to Meta researchers, could involve having two separate agents debate a topic and create the podcast outline, rather than relying on a single model to handle this aspect. NotebookLlama, like NotebookLM and other AI tools, also faces challenges with "hallucinations," meaning the generated podcasts may sometimes contain incorrect information. Use NotebookLlama. NotebookLlama aims to provide an open-source and accessible version of NotebookLM, offering several benefits to users: NotebookLlama uses Jupyter notebooks to guide users through each step of creating a podcast from a text file. Here's a simplified look at the steps involved: NotebookLlama is still in development, and there are areas where the project can improve. Enhancing the quality of the text-to-speech models could greatly improve the natural sound of generated podcasts. Future iterations could also explore different approaches, such as using multiple agents to create more engaging content. Despite these limitations, NotebookLlama provides a unique, open-source way to turn text into audio content. The approach may also have applications beyond simple PDF conversions, offering broader possibilities for creators interested in experimenting with automated text-to-speech workflows. NotebookLlama could become a valuable tool for those seeking to automate podcast creation or experiment with new forms of text-to-speech content.

Meta Takes on Google With This Open Source AI Podcast Generator

The transcription is then processed via a text-to-speech model Meta released a new open-source artificial intelligence (AI) tool on Sunday that will take on the Google NotebookLM. Dubbed NotebookLlama, the tool is an AI-powered podcast generator where users can upload a PDF file and the tool will turn it into an audio podcast with two AI characters. The tool uses three different Llama 3.1 AI models to complete the entire process. Just like Google's tool, NotebookLlama's podcast also follows a back-and-forth conversation between two AI hosts in a free-flowing manner. The Meta NotebookLlama AI tool uses three large language models to generate audio podcasts from blocks of text. Currently, the tool only accepts PDF files as input, so users will have to convert whatever text format they have into PDF. NotebookLlama first uses Llama 3.2 1B instruct model to pre-process the PDF file and save it in a '.txt' file. Then the Llama 3.1 70B instruct model is used to write a podcast transcript using the source dataset. The transcription is then dramatised using a re-writer which uses the Llama 3.1 8B instruct model. Finally, a custom tool is used to add the transcription in a text-to-speech workflow. For this, Meta is using the Parler TTS tool. Interested individuals can access all the models required to generate podcasts from the GitHub listing here. However, the AI models mentioned above are just recommendations from the developers. Users can prefer to use smaller models for every step, however, the results may vary. Meta highlighted that to run the AI system in the recommended setup, users will require a GPU with an aggregated memory of approximately 140GB. An X (formerly known as Twitter) user posted a sample of the generated podcast. Based on this, it appears the audio quality is not as good as the Google NotebookLM, and it sounds shrill and robotic. Further, there are instances where parts of audio get skipped over and the AI hosts end up speaking over each other. Meta acknowledges some of the issues and plans to improve them in the next iteration of the AI product. The company highlighted, "The TTS model is the limitation of how natural this will sound. This probably be improved with a better pipeline and with the help of someone more knowledgeable." The tech giant is also planning to use two different LLMs to write the script, where each model will debate the other to make the podcast sound more conversational. This is also part of the developers' future pipeline. Additionally, the company is also testing the Llama 405B AI model to write the transcripts as well as increasing the support for more input and output formats.

Meta releases an 'open' version of Google's podcast generator

Meta has released an "open" implementation of the viral generate-a-podcast feature in Google's NotebookLM. Called NotebookLlama, the project uses Meta's own Llama models for much of the processing, unsurprisingly. Like NotebookLM, it can generate back-and-forth, podcast-style digests of text files uploaded to it. NotebookLlama first creates a transcript from a file -- e.g. a PDF of a news article or blog post. Then, it adds "more dramatization" and interruptions before feeding the transcript to open text-to-speech models. The results don't sound nearly as good as NotebookLM. In the NotebookLlama samples I've listened to, the voices have a very obviously robotic quality to them, and tend to talk over each other at odd points. But the Meta researchers behind the project say that the quality could be improved with stronger models. "The text-to-speech model is the limitation of how natural this will sound," they wrote on NotebookLlama's GitHub page. "[Also,] another approach of writing the podcast would be having two agents debate the topic of interest and write the podcast outline. Right now we use a single model to write the podcast outline." NotebookLlama isn't the first attempt to replicate NotebookLM's podcast feature. Some projects have had more success than others. But none -- not even NotebookLM itself -- have managed to solve the hallucination problem that dogs all AI. That is to say, AI-generated podcasts are bound to contain some made-up stuff.

Meta Launches NotebookLlama, an Open Source Alternative to Google's NotebookLM

Like NotebookLM, it generates conversational, podcast-style summaries from uploaded text files. Meta has released an open source equivalent of Google's viral podcast tool, NotebookLM, called NotebookLlama. The project is built on Meta's Llama models. Like NotebookLM, it generates conversational, podcast-style summaries from uploaded text files. Check out the GitHub repository here. While people believe the quality is more robotic and doesn't sound as natural as NotebookLM, it provides insights into the techstack behind technologies like NotebookLM, giving developers the flexibility to work on it. Another open source alternative of NotebookLM includes Open NotebookLM. It is built on Llama 3.1 405B along with Fireworks AI and Instructor. As per details shared on Github, the process involves the following steps: PDF Pre-processing: Using the Llama-3.2-1B-Instruct model, the workflow first processes any PDF document, removing extra characters to ensure a clean, accurate text file. Transcript Writing: Next, the Llama-3.1-70B-Instruct model (or optionally the Llama-3.1-8B-Instruct) generates a podcast-friendly transcript from the processed text. This model adds creative touches to make the content more engaging. Dramatic Rewrite: To amplify the narrative, the Llama-3.1-8B-Instruct model rephrases the transcript for added drama, infusing personality and depth to captivate listeners. Text-to-Speech Conversion: Finally, using Parler-TTS Mini V1 and Bark's Suno, the workflow brings the text to life with a realistic, conversational audio output, ideal for podcasts. Interestingly, Llama has reached 400 million downloads globally, with India one of the top markets. With Llama 4 expected next year, Meta's Llama models are touted to be the global standard in AI -- hailed by Mark Zuckerberg as AI's "Linux moment." Top leaders like Jensen Huang and Mukesh Ambani believe these models are already building the foundation for India.