✨ Lumina AI: Multimodal Note Intelligence Hub Lumina AI is a high-performance, hackathon-winning AI Note Summarizer built with the MERN stack and powered by the Google Gemini 2.5 Flash model. It transforms messy input—handwritten notes, voice recordings, and long documents—into structured, actionable intelligence in seconds.

🚀 Core Features 🎙️ Voice-to-Summary: Real-time transcription using the Web Speech API. Turn 10-minute meetings into 5 bullet points instantly.

📸 Snap-to-Action (OCR): Upload images of whiteboards or handwritten notebooks. Gemini 2.5 handles the OCR and summarizes the visual data.

🧠 Deep Thinking Mode: Leverages the Gemini 2.5 Flash reasoning core to analyze complex notes before generating summaries.

🎭 Tone Architect: Toggle between Professional, Gen-Z (Slang), Executive Brief, and Kid-Friendly summaries.

📋 Smart Task Dashboard: Automatically extracts action items and renders them as an interactive checklist.

💬 "Ask My Notes": A contextual chat interface to ask follow-up questions about your uploaded content.

🌗 Modern UI/UX: Responsive design with Glassmorphism, Dark Mode, and smooth Tailwind animations.

🛠️ Tech Stack Frontend: React.js (Vite), Tailwind CSS, Lucide Icons, Framer Motion.

Backend: Node.js, Express.js.

AI Engine: Google Generative AI SDK (Gemini 2.5 Flash).

Deployment: Vercel (Frontend) & Railway (Backend).

🏁 Getting Started

1. Prerequisites Get a Gemini API Key from Google AI Studio.

Install Node.js.

1. Installation Bash # Clone the repo git clone https://github.com/yourusername/lumina-ai.git

Setup Backend

cd server npm install

Create .env and add: GEMINI_API_KEY=your_key_here

Setup Frontend

cd ../client npm install npm run dev 💡 Why Gemini 2.5 Flash? We chose Gemini 2.5 Flash over other models because of its native multimodality and low latency. It allows us to process images and audio without separate OCR or transcription services, reducing API costs and improving response speed for a "live" feel.

Built With

• express.js
• gemini
• node.js
• react