🚀 AstroLynx: MOSDAC AI Assistant

AstroLynx is an intelligent, conversational AI assistant for MOSDAC (Meteorological and Oceanographic Satellite Data Archival Centre). It leverages a state-of-the-art RAG pipeline, a knowledge graph, and a real-time, continuous voice chat interface to provide an engaging and powerful user experience.

---

✨ Key Features

• Advanced RAG Pipeline: Delivers accurate, context-aware answers by combining multiple retrieval strategies:
  • Hybrid Search: Fuses results from vector search (Pinecone) and a knowledge graph (Neo4j).
  • Query Transformation: Rewrites user questions and generates hypothetical documents (HyDE) to retrieve the most relevant information.
  • Intelligent Re-Ranking: Uses Reciprocal Rank Fusion (RRF) to merge and rank results from parallel searches for optimal relevance.
• Speech-to-Speech (STS) Interface: Enables hands-free, continuous conversation.
  • Voice-Activated: Automatically detects when you stop speaking to process your query.
  • Real-time STT/TTS: Transcribes your voice to text, gets a response, and plays it back as audio using Sarvam AI.
• Multi-Language Support: Seamlessly converse in multiple languages, including English, Hindi, and Hinglish. AstroLynx automatically translates, processes, and responds in your chosen language.
• Efficient & Logical Routing: Saves time and compute by intelligently classifying queries. It handles simple greetings and general questions directly, engaging the complex RAG pipeline only for MOSDAC-related topics.
• Persistent Conversations: Uses MongoDB to store chat history, allowing you to pick up conversations right where you left off.
• LangGraph Orchestration: The entire AI workflow is managed by LangGraph, ensuring a robust and logical flow between retrieval, generation, and other services.

---

🛠️ Tech Stack

Frontend:

• React & Next.js
• TypeScript
• Tailwind CSS
• Framer Motion for animations
• Lucide React for icons

Backend:

• Node.js with Express
• LangChain.js for AI orchestration, featuring:
  • LangGraph for building stateful, multi-agent applications.
  • Google Generative AI (Gemini) for language models.
  • Ollama for running local embedding models (nomic-embed-text).
  • Pinecone for vector storage and search.
  • Neo4j for knowledge graph storage and queries.
  • MongoDB for chat history and session management.
• Sarvam AI for high-quality Text-to-Speech (TTS).

Deployment & Tooling:

• Docker for containerizing services like Ollama.
• dotenv for managing environment variables.
• pnpm for frontend package management.

---

⚡ Prerequisites

• Node.js (v18 or newer)
• A MongoDB Atlas cluster or a local MongoDB instance.
• A Pinecone account and API key.
• A Neo4j AuraDB instance or local installation.
• A Google Cloud Project with the Gemini API enabled.
• A Sarvam AI API key.
• Docker (required for running Ollama).

---

🪐 Setup & Installation

1️⃣ Clone the Repository

git clone https://github.com/Dev-kaif/AstroLynx.git
cd AstroLynx

2️⃣ Configure Environment Variables

Create a .env file in the Backend directory and populate it with your credentials:

# Google Gemini API Key
GEMINI_API_KEY="your_gemini_api_key"

# Pinecone Credentials
PINECONE_API_KEY="your_pinecone_api_key"

# MongoDB Connection URI
MONGO_URI="your_mongodb_connection_string"

# Neo4j Credentials
NEO4J_URI="your_neo4j_bolt_uri"
NEO4J_USERNAME="neo4j"
NEO4J_PASSWORD="your_neo4j_password"

# Sarvam AI API Key
SARVAM_AI_API_KEY="your_sarvam_api_key"

# Ollama URL (if running locally)
OLLAMA_BASE_URL="http://localhost:11434"

3️⃣ Set Up the Backend

# Navigate to the backend directory
cd Backend

# Install dependencies
npm install

# (Required) Start Ollama and pull the embedding model
docker run -d --rm -p 11434:11434 --name ollama ollama/ollama
docker exec ollama ollama pull nomic-embed-text

# Run the backend server
npm run dev

The backend server will start on its configured port (e.g., 5000).

4️⃣ Set Up the Frontend

# Navigate to the frontend directory from the root
cd Frontend

# Install dependencies
pnpm install

# Run the development server
pnpm dev

The frontend will be available at http://localhost:3000.

---

🛰️ Future Enhancements

• Integrate a Dedicated VAD: Implement a client-side Voice Activity Detection library for more precise and responsive turn-taking in audio conversations.
• Streaming Responses: Stream text from the LLM word-by-word for a more immediate response.
• Advanced Audio Controls: Add options for speaker selection and TTS parameter customization.
• User Authentication: Implement user accounts for persistent chat history across devices.
• Expanded Knowledge Base: Ingest more domain-specific datasets for even greater expertise.

---

❤️ Contributing

Contributions are welcome! If you'd like to improve AstroLynx:

1. Fork the repository.
2. Create a new feature branch (git checkout -b feature/your-feature-name).
3. Commit your changes with clear, descriptive messages.
4. Push to the branch (git push origin feature/your-feature-name).
5. Open a Pull Request.