Speak your tasks. AI handles the rest.
A full-stack voice-to-task management system that converts natural speech into structured, prioritized tasks using GPT-4 and AssemblyAI — built by an engineer with 9 years at Nuance building enterprise voice/speech systems.
- 🎤 In-Browser Voice Recording — Record tasks naturally, no app install needed
- 🧠 AI Task Extraction — GPT-4 parses speech into structured tasks with priorities, categories, and due dates
- 🗣️ AssemblyAI Transcription — Production-grade speech-to-text pipeline
- 📋 Smart Task Management — Priorities, categories, due dates, and status tracking
- 🔐 Firebase Authentication — Secure user sessions out of the box
- ⚡ Real-Time Updates — WebSocket-powered live task syncing
- 🌐 RESTful API — Clean backend API for extensibility
- 📱 Responsive Design — Works across desktop and mobile
┌─────────────┐ WebSocket ┌──────────────────┐ ┌─────────────┐
│ React/TS │◄───────────────►│ Node.js/Express │────►│ Supabase │
│ Frontend │ REST API │ Backend API │ │ PostgreSQL │
└──────┬──────┘ └────────┬──────────┘ └─────────────┘
│ │
│ Audio ┌───────┴───────┐
└──────────────────────► │ AssemblyAI │
│ + GPT-4 │
└───────────────┘
| Layer | Technology |
|---|---|
| Frontend | React, TypeScript |
| Backend | Node.js, Express |
| Database | Supabase (PostgreSQL) |
| AI/NLP | OpenAI GPT-4 |
| Speech-to-Text | AssemblyAI |
| Auth | Firebase |
| Real-time | WebSockets |
| Deployment | Vercel |
# Clone the repo
git clone https://github.com/gaganthakur04/Narad-VoiceAI.git
cd Narad-VoiceAI
# Install dependencies
cd backend && npm install
cd ../frontend && npm install
# Configure environment
cp .env.example .env
# Add your API keys: OpenAI, AssemblyAI, Supabase, Firebase
# Run database migrations
npm run migrate
# Start development
cd backend && npm run dev
cd ../frontend && npm startScreenshots coming soon
Narad-VoiceAI/
├── backend/ # Express API server
├── frontend/ # React TypeScript app
├── supabase/ # Database migrations
└── 109 files # Production-ready codebase
Gagan Thakur — 15 years in enterprise AI, ex-Microsoft, ex-Nuance (9 years building voice & speech systems at scale).
MIT