An interactive conversational AI platform that enables users to engage in insurance consultations with AI insurance experts for India and Singapore. Built with ElevenLabs Conversational AI SDK, this application features multiple AI insurance advisors with distinct insurance packages and expertise areas.

• Single Interface Mode: Clean, professional dashboard interface for insurance consultations
• Insurance Consultation: Get personalized investment advice and insurance insights
• Real-time Voice Conversation: Powered by ElevenLabs' advanced voice AI technology
• Video Avatars: High-quality video representations of each insurance expert
• Speaking Indicators: Visual feedback showing when AI experts are responding
• Responsive Design: Works seamlessly across desktop and mobile devices

• Akshat: Singapore Expert - Specialized on Singapore insurance area

• Vanilla JavaScript with modern ES6+ features
• Webpack for module bundling and development server
• CSS3 with responsive glassmorphic design patterns
• Multi-language Support with Unicode text rendering
• HTML5 with semantic structure and accessibility features

• Node.js/Express server for API endpoints with language routing
• ElevenLabs API integration for multi-language conversational AI
• Environment-based Configuration for secure API key management
• CORS enabled for cross-origin requests

• Real-time Voice Processing with ElevenLabs SDK
• Language-specific Agent Routing based on user selection
• Cultural Context Awareness in AI responses

Before running this application, ensure you have:

• Node.js (version 18 or higher)
• ElevenLabs API Key - Get your key here
• Git for cloning the repository

git clone https://github.com/Finance-LLMs/Conversational-Insurance-Agents.git
cd Conversational-Insurance-Agents

Create a .env file in the root directory and configure your API keys:

# Windows
notepad .env

# Linux/macOS
vim .env

Add the following environment variables:

API_KEY=`your_api_key`

# Country-specific Agent
SINGAPORE_AGENT_ID=`singapore_agent_id`     # Akshat - Singapore

Frontend Dependencies:

npm install

Option A: Node.js Backend (Recommended)

npm start

Option B: Development Mode with Hot Reload

npm run dev

Open your browser and navigate to:

http://localhost:3000

1. Access the Dashboard: Open the application to see the Conversation Insurance Agents
2. Select Advisor: Choose the available AI insurance advisor (Akshat for Singapore)
3. Start Consultation: Click "Start Insurance Consultation" to begin your voice conversation
4. Insurance Discussion: Ask questions about different insurance policies, the evolving area of insurance or seek insurance advice
5. End Session: Click "End Consultation" when finished

Conversational-Insurance-Agents/
├── backend/
│   └── server.js           # Express.js server with multi-language routing
├── src/
│   ├── index.html          # Main dashboard with language selection
│   ├── app.js              # Core application logic
│   ├── styles.css          # Global styles with responsive design
│   └── videos/             # Expert video avatars
│       └── akshat.mp4      # Akshat video
├── dist/                   # Webpack build output
├── .env                    # Environment variables (not committed)
├── .gitignore              # Git ignore rules
├── package.json            # Node.js dependencies and scripts
├── webpack.config.js       # Webpack configuration
└── README.md               # This file

• npm start - Build and run production server
• npm run dev - Start development server with hot reload
• npm run build - Build for production
• npm run start:backend - Start only the Node.js backend

1. Follow modern JavaScript ES6+ standards
2. Maintain responsive design principles across all screen sizes
3. Ensure multi-language support and Unicode text handling
4. Test across different browsers, devices, and language settings
5. Maintain SEBI compliance and educational disclaimers
6. Ensure API key security and never commit secrets to version control
7. Follow cultural sensitivity guidelines for multi-language content

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