AI Agricultural Advisor

A multilingual AI advisory system for farmers in India, built with Next.js 14+, featuring voice input, image analysis, and AI-powered agricultural guidance.

Features

• 11 Indian Languages: Hindi, Bengali, English, Gujarati, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil, Telugu
• No-Login Experience: Automatic session management with UUID stored in LocalStorage
• Hybrid Audio Strategy:
  • Browser native TTS for UI elements (fast & free)
  • Sarvam AI TTS for chat responses (high-quality, natural voice)
• Multi-Modal Input:
  • Text input
  • Voice recognition (Web Speech API)
  • Image upload for plant disease detection
• Weather Integration: Real-time weather data contextual to farming advice
• Plant Health Analysis: AI-powered disease detection using Kindwise API
• PWA Support: Installable on mobile devices

Tech Stack

• Framework: Next.js 16.0.10 (App Router)
• Styling: Tailwind CSS 4
• State Management: Zustand
• Database: Neon Postgres + Drizzle ORM
• AI Services:
  • Sarvam AI (Chat + TTS)
  • Kindwise (Plant.id & Crop.health)
  • OpenWeather API
• Speech: React Speech Recognition
• Icons: Lucide React

Prerequisites

• Node.js 18+
• npm or yarn
• API Keys (required):
  • Neon Postgres Database
  • Sarvam AI API Key
  • Kindwise API Keys (Plant.id & Crop.health)
  • OpenWeather API Key

Setup Instructions

1. Clone and Install

git clone <repository-url>
cd farmer
npm install

2. Environment Variables

Create a .env.local file in the root directory with the following:

# Database (Neon Postgres)
DATABASE_URL="postgresql://..."
POSTGRES_URL="postgresql://..."
POSTGRES_URL_NON_POOLING="postgresql://..."

# API Keys
SARVAM_API_KEY="your_sarvam_api_key"
PLANT_ID_API_KEY="your_plant_id_api_key"
CROP_HEALTH_API_KEY="your_crop_health_api_key"
OPENWEATHER_API_KEY="your_openweather_api_key"

# Optional
NEXT_PUBLIC_APP_URL="http://localhost:3000"

3. Database Setup

Push the schema to your Neon database:

npm run db:push

4. Development

Run the development server:

npm run dev

Open http://localhost:3000 in your browser.

5. Build for Production

npm run build
npm start

Project Structure

farmer/
├── actions/          # Server Actions (API integrations)
│   ├── weather.ts
│   ├── analyze-image.ts
│   ├── sarvam-chat.ts
│   └── sarvam-tts.ts
├── app/              # Next.js App Router
│   ├── layout.tsx
│   ├── page.tsx
│   └── globals.css
├── components/       # React Components
│   ├── Chat/
│   │   ├── ChatInterface.tsx
│   │   ├── MessageBubble.tsx
│   │   └── InputArea.tsx
│   ├── ui/
│   │   └── SpeakingButton.tsx
│   ├── LanguageSelector.tsx
│   └── WeatherWidget.tsx
├── drizzle/          # Database Schema
│   └── schema.ts
├── hooks/            # Custom React Hooks
│   └── useChat.ts
├── lib/              # Utilities
│   ├── audio.ts
│   ├── db.ts
│   └── languages.ts
├── store/            # Zustand State
│   └── useStore.ts
└── public/           # Static Assets
    └── manifest.json

Key Features Implementation

Hybrid Audio Strategy

UI Elements (Weather, Buttons):

import { speakNative } from "@/lib/audio";
speakNative("Hello farmer", "hi-IN");

AI Chat Responses:

• Text generated by Sarvam Chat API
• Audio generated by Sarvam TTS API
• Stored as base64 and auto-played in MessageBubble

Voice Input

Uses react-speech-recognition with browser's native Web Speech API:

const { transcript, listening } = useSpeechRecognition();
SpeechRecognition.startListening({ language: "hi-IN" });

Image Analysis

Upload plant images for disease detection:

• Processed by Kindwise Plant.id or Crop.health API
• Results include disease name, probability, treatment, symptoms

Weather Context

Fetches local weather and includes in LLM prompt for contextual advice.

API Documentation

• Sarvam AI: See sarvam.md
• Kindwise: See kindwise.md

Database Schema

chats table:

• id (UUID)
• session_id (Text)
• role (user | assistant)
• content (Text)
• audio_base64 (Text, optional)
• image_url (Text, optional)
• timestamp (Timestamp)

user_settings table:

• session_id (Text, PK)
• language (Text)
• lat (Text)
• lon (Text)

Scripts

npm run dev          # Development server
npm run build        # Production build
npm run start        # Production server
npm run lint         # ESLint
npm run db:push      # Push schema to database
npm run db:studio    # Open Drizzle Studio

Deployment

Vercel (Recommended)

1. Connect your repository to Vercel
2. Add environment variables in Vercel dashboard
3. Deploy

Other Platforms

The app can be deployed to any platform that supports Next.js:

• Netlify
• AWS Amplify
• Railway
• Render

Browser Support

• Chrome/Edge: Full support (recommended)
• Firefox: Full support
• Safari: Limited speech recognition support
• Mobile browsers: Install as PWA for best experience

Troubleshooting

Speech Recognition Not Working

• Ensure you're using HTTPS (required for microphone access)
• Check browser compatibility
• Grant microphone permissions

API Errors

• Verify all API keys in .env.local
• Check API rate limits
• Ensure database connection string is correct

Build Errors

• Clear .next folder: rm -rf .next
• Reinstall dependencies: npm install
• Check Node.js version (18+)

License

MIT

Contributors

Built with ❤️ for Indian farmers