🚨 RescueAI — ریسکیو اے آئی

AI-powered emergency response platform — built for Pakistan

🌐 Live Demo · 📡 API Docs

What Is RescueAI?

RescueAI is a bilingual emergency response platform that lets anyone in Pakistan trigger an SOS alert, report an incident, track live responder dispatch, and get real-time AI guidance all in English or Urdu.

The core idea: in a real emergency, you don't have time to navigate a government portal or wait on hold. One tap sends your GPS location, assigns a responder unit, and puts an AI assistant in your hands that can walk you through CPR, bleeding control, or any crisis: by voice if needed.

Built for Pakistan. Works in Urdu. Speaks back to you.

Features

• Emergency SOS: one-tap alert with GPS coordinates, auto-assigned responder unit, and live ETA
• Incident Reporting: structured report submission with AI-generated response protocol per incident type
• Live Tracking: real-time dispatch monitoring by Alert ID or Report ID
• ARIA Assistant: conversational AI emergency guide for CPR, choking, bleeding, burns, and more
• Voice Input: browser speech recognition for hands-free operation in English and Urdu
• Text-to-Speech: ARIA responses read aloud (English via Web Speech API, Urdu via gTTS)
• Bilingual UI: full English and Urdu support with RTL layout switching

Architecture

RescueAI uses a modern, decoupled architecture to ensure 100% uptime and zero-latency during emergencies:

graph TD
    User((User)) -->|Interacts| Frontend[React 19 Frontend]
    Frontend -->|API Calls| Backend[FastAPI Backend]

    subgraph "AI & Speech Engine"
        Backend -->|Logic| Groq[Groq LLaMA 3.3 70B]
        Backend -->|Voice STT| Whisper[Groq Whisper v3]
        Backend -->|Urdu TTS| gTTS[Google gTTS]
    end

    subgraph "Maps & Data"
        Frontend -->|Maps| Leaflet[Leaflet.js]
        Frontend -->|Location| Nominatim[OpenStreetMap]
    end

Loading

AI Response Pipeline

When you submit an emergency report, the system processes it through a specialized AI pipeline to give you immediate, life-saving advice:

sequenceDiagram
    participant U as User
    participant F as Frontend
    participant B as Backend
    participant A as Groq AI

    U->>F: Submit Report
    F->>B: POST /report (Incident Details)
    B->>A: Process with Emergency Context
    A-->>B: Return Structured Protocol (EN/UR)
    B-->>F: Deliver Guidance
    F->>U: Display Guidance + Auto-Play Voice

Loading

Tech Stack

Layer	Technology	Why
Intelligence	Groq LLaMA 3.3 70B	Fast inference, strong multilingual support
Voice Control	Groq Whisper v3	Best Urdu transcription accuracy
Speech (Urdu)	Google gTTS	Correct Urdu script rendering
Backend	FastAPI	High-performance, async Python server
Frontend	React 19 + Vite	Modern, snappy user interface
Design	Tailwind CSS v4	Clean UI with RTL (Urdu) support

Project Structure

RescueAI/
├── backend/            # FastAPI app & AI logic
├── frontend/           # React 19 & Tailwind v4
├── screenshots/        # Project images (User Uploaded)
├── .gitignore          # Security & cleanup rules
└── README.md

Local Setup

Backend

1. Go to backend/
2. pip install -r requirements.txt
3. Add your GROQ_API_KEY to .env
4. Run: python main.py

Frontend

1. Go to frontend/
2. npm install
3. Run: npm run dev

License

Distributed under the MIT License. See LICENSE for more information.

Contact

Muhammad Taha Nasir - m.tahanasir.cs@gmail.com

Project Link: https://github.com/MuhammadTahaNasir/RescueAI

_{Built for Pakistan · AI Emergency Response · Bilingual · Voice-first}