Gyandex

🎯 Project Vision

Transform how we consume online content by providing flexible, AI-powered tools that convert text content into various formats. Currently focused on converting web articles and YouTube videos into podcast-ready audio content.

✨ Key Features

• Web article and YouTube video content extraction
• AI-powered conversion into natural-sounding podcast scripts
• High-quality text-to-speech synthesis with multiple voices
• Podcast feed generation for easy consumption
• Support for multiple content sources for a single podcast

🚀 Getting Started

Prerequisites

• Python 3.11-3.13
• Poetry for dependency management
• API keys for:
  • Google AI (for content processing)
  • Google Cloud (for text-to-speech)
  • S3-compatible storage (e.g., AWS S3, Cloudflare R2)

Installation

1. Clone the repository:

   git clone https://github.com/dhruvbaldawa/gyandex.git
   cd gyandex

2. Install dependencies using Poetry:

   poetry install

3. Copy .env.example to .env and set up your environment variables:

   cp .env.example .env

   Then edit .env with your API keys:

   GOOGLE_API_KEY=your_google_ai_key
   GOOGLE_CLOUD_PROJECT=your_gcp_project
   ACCESS_KEY_ID=your_s3_access_key
   SECRET_ACCESS_KEY=your_s3_secret_key

Configuration

Create a YAML configuration file (see samples/ directory for examples) with:

• Content source (YouTube URL or web article)
• Workflow settings (AI providers and models)
• Text-to-speech voice configurations
• Storage settings
• Podcast feed metadata

Usage

Generate a podcast from your configuration:

poetry run podgen your-config.yaml

📖 Documentation

Check the samples/ directory for example configurations and common use cases.

🗺️ Roadmap

• Support for PDFs and other document formats
• Frontend for easy podcast creation and management
• Enhanced customization options for podcast creation
• Question-answering interface for content interaction

License

This project is licensed under the AGPL v3 for open-source use. For those wishing to use the software in proprietary applications without disclosing source code, a commercial license is available.

• Open-Source License: AGPL v3
• Commercial License: Contact us for more information.

By contributing to this repository, you agree that your contributions will be licensed under the same AGPL v3 license.