A Telegram bot that helps farmers gets instant crop diagnosis by uploading photo of defected crop & gives smart farming suggestion by sending text & voice message. Demo

✨ Features

1. Reliable Crop Disease Diagnosis

   This system uses a fine-tuned Vision Transformer (ViT) model (google/vit-base-patch16-224) to detect diseases in rice, corn, and wheat from crop images.

   The model can identify the following diseases:

   • Rice : Bacterial Blight, Black Point, Blast, and Blight
   • Corn : Brown Spot, Common Rust, Fusarium Foot Rot, and Gray Leaf Spot
   • Wheat : Leaf Blight, Tungro, and Wheat Blast

   Fine Tuned Model : Hhsjsnns/Rice-Wheat-Corn-DiseaseCLS

2. Smart Farming Advice

   The system uses the Groq LLM to provide farming advice and disease prevention tips based on the detected crop disease.

3. Voice Support

   Farmers can ask questions using their voice. Whisper ASR converts speech into text, the Groq LLM generates a response, and gTTS converts the response back into speech.

4. Conversation Memory

   The system stores previous conversations in a PostgreSQL database. This helps the AI remember the conversation and provide more relevant and personalized responses.

5. Multilingual Support

   The system supports both Hindi and English, making it easier for farmers to interact in their preferred language.

🔄 Workflow

flowchart TD

    U[👨‍🌾 User]

    U -->|Text Query| T[Text Input]
    U -->|Crop Image| I[Image Input]
    U -->|Voice Message| V[Voice Input]

    V --> W[Whisper ASR<br/>Speech-to-Text]
    I --> M[Fine-tuned VIT Model<br/>Crop Disease Detection]

    T --> PB
    W --> PB
    M --> PB

    MEM[(Conversation Memory
    PostgreSQL)]
    MEM --> PB

    PB[Prompt Builder]

    PB--> LLM[Groq LLM]

    LLM --> A[Farming Advice & Recommendations]

    A --> TXT[📄 Text Response]
    A --> TTS[gTTS<br/>Text-to-Speech]

    TTS --> AUDIO[🔊 Voice Response]

Loading