General Introduction
Inspired by the podcast generation features of Notebook LM and the recent Open Notebook LM open source implementation. In this recipe, we will implement a detailed step-by-step guide on how to build a PDF to podcast pipeline.
Given any PDF, we will generate a dialog where the host and guest discuss and explain the content of the PDF.
In doing so, we will learn the following:
How to extract podcast scripts from PDFs using JSON schema and structured generation using open source models such as Llama 3 70b.
How to use the TTS model to bring this script to life as a dialog.
Together AI is a collection of open source projects designed to help developers build a variety of applications using Together AI. PDF to Podcast is one of the utilities that demonstrates how to convert PDF files into podcasts. With this tool, users can easily convert the content of the document into audio format, easy to listen to anytime, anywhere.
Function List
- Convert PDF files to podcasts
- Supports text conversion in multiple languages
- Provides easy-to-use examples of Jupyter Notebook
- Integration of Together AI's open source models
- Support for customized audio output formats
code example
Code: https://github.com/togethercomputer/together-cookbook/blob/main/PDF_to_Podcast.ipynb
colab: https://colab.research.google.com/github/togethercomputer/together-cookbook/blob/main/PDF_to_Podcast.ipynb
