AI Content to Chat Video

A Streamlit application that transforms AI-related content into engaging chat conversations and animated videos.

Features

• Research Agent: Analyzes and summarizes AI-related content
• Content Creation Agent: Transforms research into natural chat conversations
• Video Generation: Creates animated chat videos with customizable settings
• User-friendly Interface: Easy-to-use Streamlit interface with examples and settings

Installation

1. Clone this repository:

git clone https://github.com/yourusername/create-with-flow.git
cd create-with-flow

2. Install the required dependencies:

pip install -r requirements.txt

3. Set up environment variables: Create a .env file in the root directory with the following:

GOOGLE_API_KEY=your_google_api_key

Usage

1. Start the Streamlit app:

streamlit run streamlit_app.py

2. Open your browser and navigate to the URL shown in the terminal (usually http://localhost:8501)

3. Enter AI-related content in the text area or choose one of the example topics

4. Click "Generate Content & Video" to process your input

5. View the generated conversation and video

6. Customize video settings and regenerate if desired

7. Download the video when satisfied

Troubleshooting

PyTorch Compatibility Issues

If you encounter errors related to PyTorch and event loops, try these solutions:

1. Disable audio generation using the checkbox in the sidebar settings
2. Restart the Streamlit app
3. If the issue persists, try running with a specific Python version (3.9 is recommended)
4. Make sure your environment has compatible versions of PyTorch and Streamlit

Error messages like RuntimeError: no running event loop or Tried to instantiate class '__path__._path' are typically related to PyTorch and Streamlit compatibility issues. The app includes built-in error handling to help mitigate these issues.

Configuration

You can configure the following settings:

• Contact Name: The name displayed in the chat interface
• Input Placeholder: Text shown in the input field of the chat
• Video Speed: Playback speed of the generated video

Examples

The app includes example topics to demonstrate its capabilities:

• GPT-4 Overview
• AI Research Ethics
• Diffusion Models

Folder Structure

• agents/: Contains the multi-agent system implementation
• video_generation.py: Handles the creation of chat videos
• audio_generation.py: Manages text-to-speech for video audio
• streamlit_app.py: Main application file
• app.py: Original script for command-line usage

Requirements

• Python 3.8+
• Streamlit
• Google Generative AI API
• MoviePy
• PIL
• Other dependencies listed in requirements.txt

License

MIT

Acknowledgements

• This project uses the Google Generative AI API for research and content creation
• Video generation is built with MoviePy and PIL