AI Conversational Agent

A sophisticated conversational AI bot that enables real-time voice interactions with customizable AI characters. This project combines speech processing, natural language understanding, and real-time communication to create interactive AI agents.

Features

• Real-time voice interactions using Daily's communication platform
• Speech-to-text and text-to-speech capabilities via Deepgram
• Natural language processing using OpenAI's language models
• Customizable character personalities and scenarios
• Support for different interaction modes and difficulty levels
• Kid-friendly content filtering options
• Real-time transcript generation and handling

Prerequisites

• Python 3.8+
• Redis server
• API keys for:
  • OpenAI
  • Deepgram
  • Daily

Installation

1. Clone the repository:

git clone [your-repository-url]
cd new_agents

2. Create and activate a virtual environment:

python -m venv env
source env/bin/activate  # On Windows use: env\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file with your API keys and configuration:

OPENAI_API_KEY=your_openai_key
DEEPGRAM_API_KEY=your_deepgram_key
DAILY_API_KEY=your_daily_key

Usage

Run the bot with customized parameters:

python bot.py --character_name "Assistant" \
              --character_gender "female" \
              --character_description "A helpful AI assistant" \
              --scenario "General conversation" \
              --role "Assistant" \
              --mode "chat" \
              --kid_friendly true \
              --voice "default" \
              --room_url "your_daily_room_url" \
              --level_of_difficulty "medium"

Project Structure

• bot.py: Main application file containing core bot logic
• models.py: Data models and structures
• app_main/: Core application components
• app_tts/: Text-to-speech related functionality
• requirements.txt: Project dependencies

Technical Details

The project utilizes several key components:

• Daily SDK: For real-time voice communication
• Deepgram: Handles speech-to-text and text-to-speech conversion
• OpenAI: Powers the natural language understanding and generation
• Redis: Manages state and message queuing
• Pipecat: Custom pipeline system for processing audio and text streams

Contributing

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

License

[Your chosen license]

Contact

[Your contact information]