This project transforms PDF documents into Chinese-language podcasts, creating natural conversational audio from the text content. It is designed for content creators and researchers who want to repurpose written material into an accessible audio format.

How It Works

The system leverages a large language model (Llama-3.1-405B) to process PDF content and generate conversational dialogue. This dialogue is then synthesized into an MP3 audio file using Azure OpenAI Text-to-Speech. The architecture utilizes React and Tailwind CSS for the frontend and FastAPI for the backend.

Quick Start & Requirements

• Install: Not specified, but likely involves Python package installation and frontend build steps.
• Prerequisites: Access to Llama-3.1-405B and Azure OpenAI TTS (API keys or local setup required).
• Resources: Requires significant computational resources for the LLM and TTS models.
• Links: Demo Video, [Online Address](⚡️ PodCastLM OverView)

Highlighted Details

• Inspired by Google's NotebookLM.
• Generates natural, conversational dialogue.
• Outputs audio in MP3 format.
• Uses Llama-3.1-405B and Azure OpenAI TTS.

Maintenance & Community

• Author: YOYZHANG (Twitter: @alexu19049062).
• Contributions are welcomed via issues.
• Project is sponsored by @JiongXin and @Terry Zhang.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive license allows for commercial use and integration into closed-source projects.

Limitations & Caveats

The project requires access to large, potentially proprietary AI models (Llama-3.1-405B and Azure OpenAI TTS), which may involve significant costs and setup complexity. Specific installation and deployment instructions are not detailed in the README.