AI-powered platform that helps Indian citizens discover and apply for government schemes in seconds using voice, documents, and AI.
LinguaVerse AI is a hackathon project combining 4 core AI innovations:
| Feature | Description |
|---|---|
| SchemeMatch AI | Reverse eligibility engine — tell us who you are, we find what you qualify for |
| VoiceScheme AI | Speak in Hindi, English or Hinglish — AI understands and responds in voice |
| VisualScheme AI | Infographic dashboards, comparison charts, payment timelines |
| FormFriend AI | Auto-fills government forms and generates a pre-filled PDF |
- ✅ Matches user profile against 8+ government schemes (PM-KISAN, Ayushman Bharat, KCC, etc.)
- 🎤 Voice input using Web Speech API (Hindi/English/Hinglish)
- 📄 Document OCR simulation (Aadhaar, income certificate)
- 📊 Beautiful charts via Recharts (bar, pie, radar, line)
- 📝 jsPDF-based form generation with professional layout
- 🔗 Direct links to official scheme portals
- 📱 Fully responsive dark UI with glassmorphism design
- 🌐 Multilingual labels (Hindi + English throughout)
| Route | Description |
|---|---|
/ |
Landing page with hero and feature sections |
/onboard |
User onboarding — voice, text, or document upload |
/dashboard |
Eligibility dashboard with all results |
/schemes |
Browse all schemes with filters |
/voice |
VoiceScheme AI chatbot |
/visual |
VisualScheme AI charts |
/form |
FormFriend AI – auto-fill and download |
- Next.js 14 (App Router, TypeScript)
- TailwindCSS – custom dark theme with glassmorphism
- Framer Motion – animations
- Recharts – data visualizations
- jsPDF – PDF generation
- html2canvas – shareable screenshot
- Web Speech API – browser-native voice recognition & TTS
- FastAPI (Python) – REST API
- Pydantic – data validation
- EasyOCR / Tesseract – document OCR (production)
- OpenAI Whisper – voice transcription (production)
- GPT-4 / Claude – AI responses (production)
- PostgreSQL – database (production)
cd frontend
npm install
npm run devcd backend
python -m venv venv
venv\Scripts\activate # Windows
pip install -r requirements.txt
uvicorn main:app --reload --port 8000API docs at http://localhost:8000/docs
Copy .env.example to .env.local in /frontend:
cp frontend/.env.example frontend/.env.local# Optional – AI features work in demo mode without these
OPENAI_API_KEY=sk-...
NEXT_PUBLIC_BACKEND_URL=http://localhost:8000
FIREBASE_API_KEY=...| Scheme | Category | Benefit |
|---|---|---|
| PM-KISAN | Agriculture | ₹6,000/yr |
| Ayushman Bharat | Health | ₹5L/yr |
| Kisan Credit Card | Agriculture | ₹3L loan |
| PM Vishwakarma | Skill | ₹3.15L |
| National Scholarship | Education | ₹75K/yr |
| TN BC Scholarship | Education | ₹12K/yr |
| PMAY | Housing | ₹2.5L |
| Sukanya Samriddhi | Women | 8.2% interest |
User Profile (age, income, state, occupation, category)
↓
Rule-Based Matcher (per scheme)
↓
Score = met_criteria / total_criteria × 100
↓
100% → ELIGIBLE
60-99% → ALMOST ELIGIBLE
<60% → NOT ELIGIBLE
cd frontend
npx vercel --prod# Set environment variables in dashboard
uvicorn main:app --host 0.0.0.0 --port $PORTdocker-compose up --buildBuilt for SmartDiscover Hackathon 2026 with ❤️ for 140 crore Indians.
MIT License — Free to use, modify, and distribute.