Voice-First AI Task Orchestrator
Echo Pulse is a high-performance productivity application that leverages Natural Language Processing (NLP) to convert raw voice transcripts into structured, prioritized tasks. Built with a Zero-Trust Client Architecture, it ensures all AI logic and sensitive credentials remain secure on the server.
| Layer | Technology | Purpose |
|---|---|---|
| Framework | Next.js 15+ (App Router) | Full-stack architecture with Secure Route Handlers. |
| AI Brain | Gemini 2.5 Flash | Intent extraction and JSON schema generation. |
| State | Zustand | Lightweight, reactive client-side store for real-time updates. |
| Voice API | Web Speech API | Browser-native audio capture and transcription. |
| Styling | Tailwind CSS | Modern, dark-mode Glassmorphism UI. |
This project follows strict Server-Side Sovereignty principles:
- The Vault: All sensitive API credentials (Gemini API Key) are stored as Server-side Environment Variables (
.env.local). - Secure Proxy: The client never communicates with the Gemini API directly. It hits a Next.js Route Handler, which acts as a secure intermediary.
- Data Sanitization: The server only returns a Data Transfer Object (DTO)—structured JSON containing only the task name and priority—keeping the internal logic hidden from the browser.
- Intelligent Intent Extraction: Automatically identifies if a task is
high,medium, orlowpriority based on natural speech. - Reactive UI: Instant task rendering using Zustand state management.
- Mobile-First Design: Optimized for quick, on-the-go voice entries.
- Node.js 18+
- A Google AI Studio API Key
git clone [https://github.com/your-username/echo-pulse.git](https://github.com/your-username/echo-pulse.git)
cd echo-pulse
npm install