Skip to content

Shubham-Rasal/krishiai

Repository files navigation

KrishiAI: AI-Powered Farm Management Assistant

KrishiAI is a voice-first mobile application designed to empower Indian farmers with personalized agricultural guidance. Built with accessibility in mind, it prioritizes voice interactions while supporting text-based chat as a secondary interface.

User-Centric Design

KrishiAI is specifically designed for farmers, with a focus on:

  • Simple, intuitive interface
  • Minimal technical complexity
  • Voice-first interaction
  • Support for regional languages
  • Offline functionality for core features

Core Features

🏠 Main Screen

  • Quick access dashboard
  • Important shortcuts
  • Weather updates
  • Crop calendar
  • Recent activities

🗣️ Voice Mode

  • Real-time AI conversations
  • Powered by OpenAI with farming-specific grounding
  • Supports multiple Indian languages including Kannada
  • Focused on Karnataka region and rice farming
  • Prevents AI hallucination through data grounding

💬 Chat & Analysis Mode

  • Text-based interactions
  • Advanced image analysis:
    • Crop Disease Detection: Custom ML model with 99% accuracy
    • Pesticide Analysis: OCR-based bottle content analysis
      • pH level detection
      • Usage quantity calculation
      • Target disease/pest identification
  • Historical chat records

⚙️ Settings & Profile

  • User profile management
  • Farm details
  • Language preferences
  • Data synchronization options

Knowledge Base

Our responses are enriched with data from authoritative sources:

  • 📊 Pesticide Database

    • Chemical compositions
    • Usage guidelines
    • Safety protocols
  • 📜 Government Resources

    • Current schemes
    • Subsidy information
    • Agricultural policies
  • 🌾 Agricultural Research

    • Crop rotation guides
    • Best practices
    • Seasonal recommendations

Technical Architecture

Inference Server

  • Custom model deployment
  • Multi-model orchestration
  • Decision-making module for query routing
  • Load-balanced API endpoints

Tech Stack

  • Frontend

    • React Native v0.76
    • Expo SDK v52
    • TypeScript
    • Expo Router
    • React Native Paper
  • Backend & Storage

    • Supabase
    • Local SQLite for offline capability
    • Secure data encryption
  • AI & ML

    • OpenAI integration
    • Custom disease detection model
    • OCR processing
    • Multi-language support

Getting Started

Prerequisites

  • Node.js (LTS version)
  • Expo CLI
  • iOS Simulator or Android Emulator
  • Expo Go app (for physical device testing)

Installation

  1. Clone the repository:

git clone https://github.com/yourusername/KrishiAI.git cd KrishiAI

openai realtime api for voice upstash for vector db

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors