README.md

metadata

title: Digital Galatea AI
emoji: 🎭
colorFrom: purple
colorTo: pink
sdk: docker
pinned: false
license: mit

Digital Galatea AI

Digital Galatea is a conversational AI with a dynamic emotional model. It features a web-based interface where an avatar's shape and expression change in real-time to reflect the AI's feelings, which are influenced by the conversation.

Features

• Conversational AI: Powered by DeepSeek and Inflection AI for natural and engaging conversations.
• Dynamic Emotional Model: Simulates five core emotions: Joy, Sadness, Anger, Fear, and Curiosity.
• Responsive Avatar: The AI's visual avatar changes its shape and facial expression based on its dominant emotion.
• Sentiment Analysis: Analyzes user input to dynamically update the AI's emotional state. It uses Azure Text Analytics for high accuracy when configured, with a seamless fallback to a local NLTK VADER model.
• Real-time Web Interface: Built with Flask and JavaScript, the interface polls for updates to keep the avatar and emotion bars synchronized with the AI's state.

Tech Stack

• Backend: Python, Flask
• AI & Machine Learning:
  • DeepSeek API (Reasoning/Analysis)
  • Inflection AI (Response Generation)
  • Azure Cognitive Service for Language (Text Analytics)
  • NLTK (VADER)
• Frontend: HTML, CSS, JavaScript
• Environment Management: python-dotenv

Quick Start with Docker Hub

The easiest way to run Digital Galatea is using the pre-built Docker image from Docker Hub:

docker pull codejediondockerhub/digital-galatea:latest
docker run -d \
  --name digital-galatea \
  -p 7860:7860 \
  -e DEEPSEEK_API_KEY=your_deepseek_api_key \
  -e INFLECTION_AI_API_KEY=your_inflection_api_key \
  -e AZURE_TEXT_ANALYTICS_KEY=your_azure_key \
  -e AZURE_TEXT_ANALYTICS_ENDPOINT=https://your-resource.cognitiveservices.azure.com/ \
  codejediondockerhub/digital-galatea:latest

Or using docker-compose (see docker-compose.yml):

docker-compose up -d

Access the application at http://localhost:7860

Environment Variables

Required Environment Variables

These environment variables must be set for the application to function:

Variable	Description	Where to Get It
DEEPSEEK_API_KEY	Your DeepSeek API key for reasoning and analysis	DeepSeek Platform
INFLECTION_AI_API_KEY	Your Inflection AI API key for response generation	Inflection AI

Optional Environment Variables

These environment variables enhance functionality but are not required:

Variable	Description	Default Behavior
AZURE_TEXT_ANALYTICS_KEY	Azure Text Analytics API key for enhanced sentiment analysis	Falls back to NLTK VADER if not provided
AZURE_TEXT_ANALYTICS_ENDPOINT	Azure Text Analytics endpoint URL	Required if AZURE_TEXT_ANALYTICS_KEY is set
PORT	Port number for the Flask server	Defaults to 7860

Environment Variable Examples

Using Docker Run

docker run -d \
  --name digital-galatea \
  -p 7860:7860 \
  -e DEEPSEEK_API_KEY=sk-your-deepseek-key \
  -e INFLECTION_AI_API_KEY=your-inflection-key \
  -e AZURE_TEXT_ANALYTICS_KEY=your-azure-key \
  -e AZURE_TEXT_ANALYTICS_ENDPOINT=https://your-resource.cognitiveservices.azure.com/ \
  codejediondockerhub/digital-galatea:latest

Using Environment File

Create a .env file:

# Required
DEEPSEEK_API_KEY=sk-your-deepseek-key
INFLECTION_AI_API_KEY=your-inflection-key

# Optional - Enhanced Sentiment Analysis
AZURE_TEXT_ANALYTICS_KEY=your-azure-key
AZURE_TEXT_ANALYTICS_ENDPOINT=https://your-resource.cognitiveservices.azure.com/

# Optional - Port Configuration
PORT=7860

Then run with:

docker run -d --name digital-galatea -p 7860:7860 --env-file .env codejedi/digital-galatea:latest

Docker Hub

The image is available on Docker Hub:

Image: codejediondockerhub/digital-galatea:latest

docker pull codejediondockerhub/digital-galatea:latest

Building and Pushing to Docker Hub

If you want to build and push your own version:

# Build the image
docker build -t codejediondockerhub/digital-galatea:latest .

# Login to Docker Hub
docker login

# Push to Docker Hub
docker push codejediondockerhub/digital-galatea:latest

Local Development

1. Clone the Repository

   git clone https://github.com/codejedi-ai/Digital-Galatea.git
   cd Digital-Galatea
   

2. Create and Activate a Virtual Environment

   # For Windows
   python -m venv venv
   .\venv\Scripts\activate
   
   # For macOS/Linux
   python3 -m venv venv
   source venv/bin/activate
   

3. Install Dependencies

   pip install -r requirements.txt
   

4. Configure Environment Variables

   • Copy .env.example to .env
   • Add your API keys (see Environment Variables section above)

5. Run the Application

   python app.py
   

6. Access the Web Interface

   • Open your browser and navigate to http://127.0.0.1:7860
   • The AI will initialize in the background. Once ready, you can start chatting.

Deployment Options

Heroku

The repository includes Heroku configuration files (Procfile, app.json, runtime.txt).

1. Connect your GitHub repository to Heroku
2. Set environment variables in Heroku dashboard
3. Deploy automatically on push to main branch

Hugging Face Spaces

1. Go to your Space Settings → Repository secrets
2. Add the required secrets (see Environment Variables section)
3. The app will build and deploy automatically

API Endpoints

The application exposes the following endpoints:

• GET /: Serves the main chat interface.
• POST /api/chat: Handles chat messages and returns the AI's response.
• GET /api/avatar: Provides the current avatar shape, emotions, and sentiment for real-time frontend updates.
• GET /api/is_initialized: Check if the system is initialized and ready.
• GET /status: Reports the initialization status of the AI components.
• GET /health: A simple health check endpoint.

How It Works

1. User Input: You type a message in the chat interface
2. Sentiment Analysis: The system analyzes the emotional tone of your message using Azure Text Analytics or NLTK VADER
3. Emotional Processing: Galatea's emotional state is updated based on the conversation
4. AI Response: DeepSeek analyzes the context, and Inflection AI generates a contextual response that reflects Galatea's emotional state
5. Avatar Update: The visual avatar changes shape and expression based on the dominant emotion:
   • Circle → High Joy (happy, positive state)
   • Triangle → High Sadness (melancholic state)
   • Square → Neutral or mixed emotions

Credits

This project demonstrates the integration of:

• DeepSeek API for reasoning and analysis
• Inflection AI for response generation
• Azure Text Analytics and NLTK for sentiment analysis
• Real-time emotional modeling
• Dynamic visual representation of AI state

---

Made with ❤️ for exploring emotional AI