🎙️ Real-time fact-checking for live debates

Key Features • Demo • Getting Started • How It Works • Configuration • Roadmap • FAQ • Technologies • Contributing • License

The AI-Powered Debate Fact-Checker is a cutting-edge tool designed to enhance the quality and accuracy of public discourse. By leveraging advanced AI and NLP technologies, this application provides real-time fact-checking, sentiment analysis, and topic tracking for live debates.

• Real-time Audio Processing: Transcribe live debate audio and identify speakers.
• Automated Claim Extraction: Identify and isolate fact-checkable claims from transcribed text.
• AI-Powered Fact-Checking: Utilize the Groq API for rapid, intelligent fact verification.
• Interactive Visualizations:
  • Timeline of claims and their verification status
  • Network graph of speakers and topics
  • Truth meter for sentiment analysis
• Efficient Web Search: Gather relevant information to support fact-checking with caching and rate limiting.
• Context-Aware Analysis: Maintain debate context for more accurate fact-checking and topic tracking.
• Summary Generation: Produce concise summaries of debate key points and verification results.

• Python 3.8+
• Groq API key
• Hugging Face API token (for speaker diarization)

1. Clone the repository:

   git clone https://github.com/yourusername/ai-debate-fact-checker.git
   cd ai-debate-fact-checker
   

2. Install required packages:

   pip install -r requirements.txt
   

3. Set up environment variables: Create a .env file in the project root and add your API keys:

   GROQ_API_KEY=your_groq_api_key_here
   HUGGINGFACE_TOKEN=your_huggingface_token_here
   

Run the Streamlit app:

streamlit run main.py

Navigate to the provided local URL in your web browser to access the application.

1. Audio Upload: Users upload a debate audio file (WAV format).
2. Transcription: The audio is transcribed and processed for speaker diarization.
3. Claim Extraction: AI identifies fact-checkable claims from the transcribed text.
4. Fact-Checking: Each claim is verified using web searches and AI analysis.
5. Visualization: Results are displayed with interactive charts and graphs.
6. Summary: A concise summary of the debate and fact-checking results is generated.

• Streamlit: Web application framework
• Groq API: AI-powered text generation and analysis
• spaCy: Natural Language Processing
• pyannote.audio: Speaker diarization
• Plotly: Interactive visualizations
• NetworkX: Graph creation for topic/speaker relationships
• Beautiful Soup: Web scraping for fact-checking
• aiohttp: Asynchronous HTTP requests

Contributions are welcome! Please feel free to submit a Pull Request.

This project is licensed under the MIT License - see the LICENSE file for details.

• Groq for the Groq API
• The open-source & hugging face community for the amazing tools and libraries used in this project

Built with ❤️ by [Elena J. Greer]