VAANI - AI Voice Assistant for Government Services by Roshan Kumar SinghVAANI - AI Voice Assistant for Government Services by Roshan Kumar Singh

VAANI - AI Voice Assistant for Government Services

Roshan Kumar Singh

Roshan Kumar Singh

๐ŸŽ™๏ธ VAANI - Voice-First Civic Engagement Platform

AI-Powered Voice Assistant for Government Services & Civic Complaints
Features โ€ข Quick Start โ€ข Architecture โ€ข Documentation โ€ข Contributing

๐Ÿ“– Overview

VAANI is a comprehensive voice-first platform designed to bridge the gap between citizens and government services. It leverages cutting-edge AI technology to provide multilingual voice assistance for filing civic complaints, tracking issues, and accessing government schemes.

๐ŸŽฏ Key Highlights

๐Ÿ—ฃ๏ธ Multilingual Support: Seamlessly handles Hindi, English, and Hinglish
๐ŸŽค Voice-First Design: Natural voice interactions powered by AWS Transcribe & Polly
๐Ÿค– AI-Powered: Intelligent responses using AWS Bedrock Nova Sonic
๐Ÿ“Š Real-Time Analytics: Interactive dashboards with live data visualization
๐Ÿ” Secure Authentication: Firebase-based user management
๐Ÿ“ฑ Fully Responsive: Optimized for mobile, tablet, and desktop
โ™ฟ Accessible: WCAG-compliant design principles

๐ŸŒŸ Live Demo

โœจ Features

๐ŸŽ™๏ธ Voice Assistant

Natural Language Processing: Understands context and intent
Multi-turn Conversations: Maintains conversation history
Language Detection: Automatically detects user's preferred language
Voice Commands: Hands-free interaction for accessibility
Real-time Transcription: Instant speech-to-text conversion
Neural Voice Synthesis: Natural-sounding responses

๐Ÿ“ Civic Engagement

Complaint Filing: Easy submission of civic issues
Issue Tracking: Real-time status updates
Category Management: Healthcare, Legal, Education, Welfare
Priority Handling: Urgent issues flagged automatically
Historical Records: Complete activity history

๐Ÿ“Š Analytics & Insights

Interactive Charts: Issue distribution and trends
Performance Metrics: Resolution rates and response times
User Activity: Comprehensive activity tracking
Admin Dashboard: System-wide analytics and monitoring

๐Ÿ”’ Security & Privacy

Firebase Authentication: Secure user management
Role-Based Access: User and Admin roles
Data Encryption: End-to-end security
Privacy Compliance: GDPR-ready architecture
๐ŸŽค Supported Voice Queries
VAANI understands natural voice queries from citizens in English and Hinglish. Below are some example questions the assistant can handle.
๐Ÿ‘‹ General Interaction
Tum kaun ho?
Vaani kya hai?
Tum kya madad kar sakte ho?
Namaste
Mujhe madad chahiye
Mai bahut pareshan hu
Mujhe kuch poochna hai
Aapka naam kya hai?
Dhanyavaad
Alvida
๐Ÿ™๏ธ Civic Issues & Complaints
Complaint kaise kare?
Complaint status kaise check kare?
Sadak tut gayi hai kya kare?
Street light kharab hai kya kare?
Pani nahi aa raha kya kare?
Kachra nahi uth raha kya kare?
Sewer overflow ho raha kya kare?
Bijli nahi aa rahi hai
๐Ÿ‘ค Account & System
Account kaise banaye?
๐Ÿ›๏ธ Government Schemes
Sarkari yojna kya hai?
Ujjwala yojna kya hai?
Ayushman card ke liye eligible kaun hai?
How to make Ayushman card?

๐Ÿš€ Quick Start

Prerequisites

Node.js 20+ (Download)
Python 3.11+ (Download)
AWS Account with Bedrock access
Firebase Project (Setup Guide)

1๏ธโƒฃ Clone Repository


2๏ธโƒฃ Frontend Setup


๐ŸŒ Frontend runs on: http://localhost:5173

3๏ธโƒฃ AI Backend Setup (Node.js)


๐Ÿš€ Backend runs on: http://localhost:5000

4๏ธโƒฃ Python Backend Setup (Optional)


๐Ÿ Python server runs on: http://localhost:8000

5๏ธโƒฃ FastAPI Backend Setup


โšก FastAPI runs on: http://localhost:8080

๐Ÿ“ Project Structure


๐Ÿ—๏ธ Architecture

System Overview


Technology Stack

Layer Technology Purpose Frontend React 18, Vite, Tailwind CSS User interface UI Components Framer Motion, Recharts, Lucide Icons Animations & visualizations Backend Node.js, Express, FastAPI API servers AI/ML AWS Bedrock Nova Sonic Natural language processing Voice AWS Transcribe, Polly Speech-to-text & text-to-speech Authentication Firebase Auth User management Storage AWS S3 Audio file storage Deployment AWS Amplify, Lambda Serverless infrastructure State Management React Context API Global state Routing React Router v7 Client-side routing

๐Ÿ“š Documentation

Core Documentation

๐Ÿ“ Architecture Guide - System design and infrastructure
๐ŸŽค Voice Assistant Guide - Voice features and integration
๐Ÿงช Voice Assistant Testing - Testing procedures
๐Ÿ“Š Project Status - Current development status
โ˜๏ธ AWS Quota Guide - AWS service limits

Backend Documentation

๐Ÿ“– AI Backend README - Node.js backend setup
๐Ÿ Python Backend README - Python backend setup
โšก FastAPI Backend README - Authentication backend

Training & Testing

๐ŸŽ“ Training Guide - AI model training
โœ… Test Summary - Test results
๐Ÿ“ Questions Guide - Supported queries

๐Ÿงช Testing

Frontend Testing


Backend Testing


Python Backend Testing


๐Ÿ”ง Configuration

Environment Variables

Frontend (.env)


AI Backend (.env)


FastAPI Backend (.env)


๐Ÿ“Š Dataset

Training Data

Active Dataset: training-dataset.json (61 conversations)
Backup Dataset: training-dataset-backup-213.json (213 conversations)
Languages: Hindi, English, Hinglish
Categories: Healthcare, Legal, Education, Welfare, Government Schemes

Dataset Statistics

โœ… Training Accuracy: 100%
โœ… Language Detection: Working
โœ… Intent Classification: Optimized
โœ… Response Cache: Active

๐ŸŽฏ Current Status

โœ… Completed Features

Voice assistant with multilingual support
User authentication and authorization
Admin and user dashboards
Interactive analytics charts
Issue tracking system
Responsive design (mobile, tablet, desktop)
AWS Amplify integration
Firebase authentication
Training data optimization (100% accuracy)
Response caching system
Language detection
Intent classification

๐Ÿšง In Progress

Real-time notifications
Advanced analytics
Multi-region deployment
Performance optimization

๐Ÿ”ฎ Planned Features

Mobile app (React Native)
Offline mode
Advanced AI models
Community forums
Video support

๐Ÿš€ Deployment

AWS Amplify Deployment


Manual Deployment

Frontend (Vercel/Netlify)


Backend (AWS Lambda)


๐Ÿค Contributing

We welcome contributions! Please follow these steps:
Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Commit changes: git commit -m 'Add amazing feature'
Push to branch: git push origin feature/amazing-feature
Open a Pull Request

Development Guidelines

Follow existing code style
Write meaningful commit messages
Add tests for new features
Update documentation
Ensure responsive design

๐Ÿ“ Scripts Reference

Frontend Scripts


AI Backend Scripts


๐Ÿ› Troubleshooting

Common Issues

Issue: Voice assistant not responding
โœ… Check AWS credentials in .env
โœ… Verify Bedrock access in AWS console
โœ… Check browser microphone permissions
Issue: Authentication failing
โœ… Verify Firebase configuration
โœ… Check serviceAccountKey.json
โœ… Ensure Firebase project is active
Issue: Charts not displaying
โœ… Check console for errors
โœ… Verify Recharts installation
โœ… Clear browser cache

๐Ÿ“„ License

This project is licensed under the MIT License - see the LICENSE file for details.

๐Ÿ‘ฅ Team

Built with โค๏ธ by the VAANI Team

๐Ÿ™ Acknowledgments

AWS for cloud infrastructure
Firebase for authentication
React community for amazing tools
Open source contributors

๐Ÿ“ž Support

For support, email vaani.voice.assist@gmail.com or join our Slack channel.
Made with ๐Ÿ’œ by VAANI Team
Like this project

Posted Mar 30, 2026

Developed a voice assistant platform enhancing citizen access to government services.