🌾 AgroGPT: Intelligent Multilingual Agri-Advisory System

Bridging the Digital Divide with Voice-Enabled Generative AI for Agriculture

---

🌟 Project Vision

AgroGPT is an enterprise-grade, multilingual AI advisor designed to empower farmers and agriculturists. In regions where literacy or technical barriers exist, AgroGPT provides a voice-first, native-language interface to access expert advice on crop management, pest control, and sustainable farming practices.

This system leverages state-of-the-art Generative AI and ASR (Automatic Speech Recognition) via the Sarvam AI platform to deliver high-accuracy, domain-specific intelligence.

---

🚀 Key Features

1. 🎤 Voice-First Interaction

Seamlessly transcribe regional dialects into actionable queries using advanced ASR models. Designed for accessibility in rural environments.

2. 🌍 Universal Multilingual Support

Conversational support for 6+ major languages:

• English, Hindi, Bengali, Gujarati, Kannada, Punjabi.

3. 🛡️ Enterprise Security & Scalability

• API Key Guard: Custom-built middleware for secure, authenticated access.
• Rate Limiting: Integrated slowapi to prevent abuse and ensure high availability.
• Async Architecture: Built on FastAPI for high-performance, asynchronous processing.

4. 📦 Cloud-Ready Deployment

• Full Dockerization for consistent environments.
• GitHub Actions (CI/CD) automated pipeline for Docker Hub deployment.

---

🛠️ Tech Stack

Component	Technology
Backend	FastAPI (Python), SQLAlchemy, Pydantic
Frontend	React.js, Vite, Tailwind CSS, Axios
AI/ML	Sarvam AI (Text-to-Text & ASR), saarika:v2.5
Database	SQLite (Production-ready abstraction available)
DevOps	Docker, Docker Compose, GitHub Actions

---

🏗️ System Architecture

User Interface (React) <---> Secure Gateway (FastAPI) <---> Sarvam AI Engine
                                     |
                          +----------+----------+
                          |          |          |
                    Local Cache   Auth Mgr   Session DB

---

📁 Project Structure

AgroGPT-Multilingual-Voice-Chat/
│
├── 📂 .github/
│   └── 📂 workflows/
│       └── docker-publish.yml       # CI/CD pipeline → auto-push to Docker Hub
│
├── 📂 backend/                      # FastAPI Python backend
│   ├── 📂 app/
│   │   ├── 📂 core/                 # Core utilities & configuration
│   │   │   ├── __init__.py
│   │   │   ├── config.py            # App settings & environment variables
│   │   │   ├── database.py          # SQLAlchemy DB session setup
│   │   │   ├── logger.py            # Centralized logging setup
│   │   │   └── security.py          # API-Key authentication middleware
│   │   │
│   │   ├── 📂 models/               # Data layer
│   │   │   ├── __init__.py
│   │   │   ├── db_models.py         # SQLAlchemy ORM models (DB tables)
│   │   │   └── schemas.py           # Pydantic request/response schemas
│   │   │
│   │   ├── 📂 routers/              # API route handlers
│   │   │   ├── __init__.py
│   │   │   └── chat.py              # /chat endpoints (text & voice)
│   │   │
│   │   ├── 📂 services/             # Business logic & external API clients
│   │   │   ├── __init__.py
│   │   │   └── sarvam_service.py    # Sarvam AI integration (LLM + ASR)
│   │   │
│   │   ├── __init__.py
│   │   └── main.py                  # FastAPI app entry point
│   │
│   ├── 📂 data/                     # SQLite database storage (runtime, not committed)
│   ├── .dockerignore
│   ├── Dockerfile                   # Backend container definition
│   └── requirements.txt             # Python dependencies
│
├── 📂 frontend/                     # React.js frontend (Vite)
│   ├── 📂 src/
│   │   ├── 📂 components/           # Reusable UI components
│   │   │   ├── ChatInterface.jsx    # Main chat UI (text + voice input)
│   │   │   └── Sidebar.jsx          # Language selector & session controls
│   │   │
│   │   ├── 📂 hooks/                # Custom React hooks
│   │   │   └── useAPI.js            # Axios API call abstraction
│   │   │
│   │   ├── App.jsx                  # Root application component
│   │   ├── main.jsx                 # React DOM entry point
│   │   └── index.css                # Global styles
│   │
│   ├── index.html                   # HTML shell
│   ├── vite.config.js               # Vite bundler configuration
│   ├── tailwind.config.js           # Tailwind CSS configuration
│   ├── postcss.config.js            # PostCSS configuration
│   ├── package.json                 # Node.js dependencies
│   ├── .dockerignore
│   └── Dockerfile                   # Frontend container definition
│
├── 📂 docs/                         # Project documentation & reports
│   ├── INTERVIEW_SCRIPT.md
│   ├── PROJECT_ARCHITECTURE_AND_INTERVIEW_GUIDE.md
│   └── Project_Report.md
│
├── docker-compose.yml               # Multi-service orchestration (backend + frontend)
├── run.py                           # One-command local dev launcher
├── .env                             # Environment variables (not committed)
├── .gitignore
└── README.md

---

⚙️ Setup & Installation

1. Environment Configuration

Create a .env file in the root directory:

SARVAM_API_KEY="your_sarvam_key"
MASTER_API_KEY="my-secret-master-key-12345"

2. Local Development (Classic)

Backend:

cd backend
pip install -r requirements.txt
uvicorn app.main:app --port 8000 --reload

Frontend:

cd frontend
npm install
npm run dev

3. Modern Launch (Recommended)

Launch the entire stack with one command:

python run.py

---

🐳 Dockerization & Deployment

To run in a containerized environment (Port 8000 & 5173):

docker-compose up --build -d

CI/CD Pipeline: The project is configured to automatically push images to Docker Hub:

• Backend: rashedulalbab1234/agrogpt-backend
• Frontend: rashedulalbab1234/agrogpt-frontend
• GitHub User: rashedulalbab253

---

🎓 Academic Significance

This project serves as a research foundation for:

1. NLP in Low-Resource Languages: Real-world application of cross-lingual knowledge transfer.
2. HCI (Human-Computer Interaction): Studying the impact of voice-based AI on rural technology adoption.
3. Domain-Specific AI Alignment: Implementing strict system prompts and safety guards for critical agricultural advice.

---

�️ Contribution & Role

As the Lead Developer of AgroGPT, I spearheaded the following architectural and technical implementations:

• System Orchestration: Designed and implemented the full-stack architecture using FastAPI and React.
• Multilingual Integration: Integrated Sarvam AI's Generative models and ASR (saarika:v2.5) to support diverse regional languages.
• Security Engineering: Built custom API-Key middleware and enforced rate-limiting to ensure system integrity.
• DevOps & Cloud: Engineered the Dockerization strategy and CI/CD pipelines for automated multi-service deployment.
• Agricultural Alignment: Configured domain-specific system prompts to ensure the AI remains focused on providing accurate agricultural advice.

---

�👨‍💻 Author

Rashedul Albab

• GitHub
• Docker Hub

---

📜 License

Licensed under the MIT License.