/research-paper-to-podcast

Yet Another Deep Dive Clone (YADDC): An automated system that transforms academic research papers into engaging podcast conversations using CrewAI and ElevenLabs.

Primary LanguageJupyter Notebook

AI Podcast Generator

An automated podcast generation system using CrewAI and ElevenLabs. The system reads research papers and creates engaging, conversational podcasts with natural-sounding voices.

🎧 Latest AI Research Paper Podcast

Spotify

Latest Episode: "Long Context vs. RAG for LLMs"

  • 🔍 Deep dive into the performance comparison
  • 🧠 Analysis of memory vs retrieval approaches
  • 💡 Real-world implications and use cases

Generated using AI agents for research analysis, script writing, and voice synthesis.

Features

  • Research paper analysis and summarization
  • Natural conversational script generation
  • Enhanced script refinement for engagement
  • High-quality voice synthesis using ElevenLabs
  • Professional audio mixing and processing

Setup

  1. Clone this repository
  2. Install requirements: 
    pip install -r requirements.txt
  3. Create a .env file with your API keys: 
    OPENAI_API_KEY=your_key_here
ELEVENLABS_API_KEY=your_key_here
ANTHROPIC_API_KEY=your_key_here

Usage

  1. Place your research paper in the knowledge/ directory or use the one provided in the repository
  2. Run the agents script: 
    python podcast_generator.py
  3. Find outputs in the outputs/ directory:
    • Generated scripts
    • Audio segments
    • Final podcast

Configuration

  • Voice settings can be adjusted in tools.py
  • Agent behaviors configured in podcast_generator.py

TODO

  • Add Docling to perform RAG on various file types
  • Add multilingual support
  • Add search functionality to reaserch supporting materials for the podcast

Acknowledgements