🏆 Adobe PDF Insights & Podcast Platform

Grand Finale Submission - Complete AI-Powered Solution

🎯 One-Command Deployment Ready for Jury Evaluation

🎉 Complete PDF Processing, AI Insights & Podcast Generation Platform

🎬 Watch Demo Video for Quick Feature Overview

---

🚀 Quick Start for Jury

One-Command Deployment (Recommended)

command 1 : docker build --platform linux/amd64 -t yourimageidentifier .
command 2 : docker run --rm \
  -v "${PWD}/credentials:/credentials" \
  -e ADOBE_EMBED_API_KEY="c0598f2728bf431baecd93928d677adc" \
  -e LLM_PROVIDER="gemini" \
  -e GOOGLE_API_KEY="api_key" \
  -e GEMINI_MODEL="gemini-model" \
  -e TTS_PROVIDER="azure" \
  -e AZURE_TTS_KEY="api_key" \
  -e AZURE_TTS_ENDPOINT="Your_endpoint" \
  -e AZURE_TTS_REGION="Your_region" \
  -p 8080:8080 \
  yourimageidentifier

System Requirements

• Docker Desktop with 4GB+ RAM allocated
• Port 8080 available (configurable in docker-compose.yml)
• Internet access for AI APIs (Google Gemini & Azure TTS)

Access Points

• 🌐 Main Application: http://localhost:8080
• 📚 API Documentation: http://localhost:8080/docs
• ❤️ Health Check: http://localhost:8080/health

---

📊 Performance Metrics

Metric	Performance	Details
📄 PDF Processing	~0.16s per document	Includes outline extraction and indexing
🔍 Search Response	<100ms indexed	TF-IDF vectorization with caching
💡 Insight Generation	2-5s per request	Google Gemini API processing time
🎙️ Audio Generation	3-8s per script	Azure TTS synthesis and encoding
🚀 Container Startup	~30-40s	Complete application initialization
💾 Memory Usage	2-4GB	Optimized for production deployment
🌐 Frontend Bundle	<2MB gzipped	Vite optimization with code splitting
🔄 API Response	<50ms average	FastAPI with async processing

🎯 Core Features Implemented

✅ Mandatory Requirements

• 📄 PDF Upload & Processing - Drag & drop with automatic outline generation
• 🔍 Advanced Semantic Search (Heading) - Cross-document content discovery with the model made in Round 1A
• 🎯 Text Highlighting - Real-time text highlighting in PDF viewer during search
• 🤝 Content Connections - Intelligent relationship discovery between documents
• ⚡ High Performance - Sub-second response times, optimized algorithms
• 🎨 Modern UI - Responsive design with smooth interactions and animations

🏆 Bonus Features (+10 Points)

• 💡 AI Insights Bulb (+5) - 5 insight types: takeaways, contradictions, examples, cross-references, facts
• 🎙️ Podcast Generation (+5) - AI-generated audio content with Azure TTS
• 🎯 PDF Navigation - Click-to-navigate with text highlighting via Adobe Embed API
• 🔄 Real-time Updates - Live connection discovery and insight caching

🌟 Advanced Features

• 📖 Inline PDF Viewer - Embedded Adobe PDF Embed API with text selection
• 🎨 Text Selection Popup - Interactive popup for selected text with connections
• 🔊 Audio Playback - Integrated audio player for generated podcasts
• 📱 Responsive Design - Works seamlessly on desktop, tablet, and mobile
• ⚡ Real-time Search - Instant search results with debounced input
• 🎭 Framer Motion Animations - Smooth transitions and micro-interactions

---

🎯 Feature Spotlight: Advanced Semantic Search with Text Highlighting

How It Works

Our intelligent semantic search system provide accurate content discovery across all uploaded PDFs. When you search for content, the system:

1. Analyzes Your Query - Processes natural language queries using advanced NLP
2. Searches Document Outlines - Finds relevant headings and sections across all PDFs
3. Ranks Results - Uses relevance scoring with exact match prioritization
4. Highlights Content - Opens the relevant document and highlights matching text
5. Shows Context - Displays the document page and surrounding content

🎯 Text Highlighting in PDF Viewer

When you click on a search result:

• Automatic Navigation - Opens the document to the exact page
• Text Highlighting - Highlights the matching content in the PDF viewer
• Adobe Embed Integration - Uses Adobe PDF Embed API for precise text highlighting
• Smooth Transitions - Framer Motion animations for seamless navigation

Search Algorithm

# Linux/macOS - Make executable and run
chmod +x run-jury.sh
./run-jury.sh

# Windows - Run batch file
run-jury.bat

# Direct Docker Compose (Alternative)
docker-compose up --build

# Silent background deployment
docker-compose up --build -d

Docker Commands

# Build only (without running)
docker build -t adobe-pdf-platform .

# Run with custom port
docker run -p 3000:8080 adobe-pdf-platform

# Run with environment variables
docker run -p 8080:8080 \
  -e GOOGLE_API_KEY=your_key \
  -e AZURE_TTS_KEY=your_key \
  adobe-pdf-platform

🛠️ Development Commands

Local Development Setup

# Backend development
cd backend
python -m venv venv
source venv/bin/activate  # Linux/Mac
# or
venv\Scripts\activate     # Windows

pip install -r requirements.txt
python main.py

# Frontend development
cd frontend
npm install
npm run dev

# Run both in development mode
npm run dev:all

Development with Docker

# Development build with hot reload
docker-compose -f docker-compose.dev.yml up

# Run specific services
docker-compose up backend
docker-compose up frontend

# Execute commands in running container
docker-compose exec adobe-pdf-platform bash
docker-compose exec adobe-pdf-platform python backend/main.py

---

🏗️ Technical Architecture

Single Container Solution

📦 Docker Container (Production-Ready)
├── 🎨 Frontend (React 19 + Vite)
│   ├── Adobe PDF Embed API Integration
│   ├── Modern UI with Framer Motion
│   ├── TailwindCSS Styling
│   ├── Real-time Text Selection
│   └── Responsive Design System
├── 🔧 Backend (FastAPI + Python 3.11)
│   ├── PDF Processing Pipeline
│   ├── Google Gemini AI Integration  
│   ├── Azure Text-to-Speech
│   ├── Round 1A Search Engine
│   ├── Connection Discovery Engine
│   └── RESTful API Endpoints
├── 🧠 AI Services
│   ├── Google Gemini (Text Analysis)
│   ├── Azure TTS (Audio Generation)
│   ├── TF-IDF Vectorization
│   └── Cosine Similarity Matching
└── 💾 Storage
    ├── PDF Files
    ├── Generated Outlines
    ├── Audio Content
    └── Search Indices

Component Architecture

Frontend Components (React 19)

src/
├── components/
│   ├── LeftPanel.jsx           # File list and search interface
│   ├── CenterPanel.jsx         # PDF viewer and content display
│   ├── RightPanel.jsx          # Insights and podcast tabs
│   ├── search/
│   │   ├── SemanticSearch.jsx  # AI-powered search interface
│   │   └── FilenameSearch.jsx  # Basic filename filtering
│   ├── InlinePDFViewer.jsx     # Adobe Embed API wrapper
│   ├── TextSelectionPopup.jsx  # Interactive text selection
│   ├── InsightsTab.jsx         # AI insights generation
│   └── PodcastTab.jsx          # Audio content generation
├── services/
│   └── api.js                  # API communication layer
└── utils/
    └── pdfDb.js               # Client-side storage utilities

Backend Services (FastAPI)

backend/
├── main.py                    # Application entry point
├── config.py                  # Environment configuration
├── api/
│   ├── documents.py           # PDF upload and management
│   ├── search.py              # Semantic search endpoints
│   ├── insights.py            # AI insights generation
│   ├── connections.py         # Document relationship discovery
│   └── podcast.py             # Audio generation endpoints
├── services/
│   ├── document_service.py    # PDF processing logic
│   ├── search_service.py      # TF-IDF search implementation
│   ├── insights_service.py    # Gemini AI integration
│   ├── connection_service.py  # Relationship analysis
│   └── podcast_service.py     # Azure TTS integration
├── models/                    # Data models and schemas
├── outline_engine/            # PDF outline extraction
└── utils/                     # Shared utilities and helpers

---

🎯 Jury Evaluation Guide

Quick Validation Steps

1. 🚀 Deploy: Run docker-compose up --build or ./run-jury.sh
2. ✅ Verify: Check http://localhost:8080/health (should return {"status": "healthy"})
3. 📄 Upload: Try uploading a PDF via the drag & drop interface
4. 🔍 Search: Test semantic search functionality with sample queries
5. 💡 Insights: Generate AI insights for selected text content
6. 🎙️ Audio: Create a podcast from uploaded PDFs
7. 🎯 Navigation: Test click-to-navigate from search results

Key Differentiators to Evaluate

• 🏆 Complete Integration - Frontend + Backend in single container with health checks
• ⚡ Production Ready - Comprehensive error handling, logging, and monitoring
• 🎨 User Experience - Modern, responsive interface with smooth animations
• 🧠 AI Integration - Multiple AI services (Google Gemini + Azure TTS) working seamlessly
• 📈 Performance - Optimized for speed, reliability, and resource efficiency
• 🔍 Advanced Search - Real-time semantic search with text highlighting and navigation
• 📖 PDF Integration - Deep Adobe PDF Embed API integration with text selection

Feature Verification Checklist

Core Requirements ✅

• PDF upload and processing pipeline works end-to-end
• Semantic search returns relevant results across documents
• Search results navigate to exact PDF pages with highlighting
• Content connection discovery finds relationships between documents
• High-performance response times (search <100ms, processing <1s)
• Modern, responsive UI works on different screen sizes

Bonus Features ✅

• AI Insights Bulb generates 5 different insight types
• Podcast generation creates audio content from PDF text
• Text selection in PDF triggers instant insight generation
• Real-time search updates as user types
• Smooth animations and micro-interactions throughout UI

Preview Command Issues

# If npm run preview fails, ensure build files exist
cd frontend
npm run build                      # Build production files first
npm run preview                    # Then preview

# Check if port 3000 is available
sudo lsof -i :3000                # Check what's using port 3000
npx kill-port 3000                # Kill process on port 3000

# Use different port for preview
npm run preview -- --port 3001    # Use port 3001 instead

# Verify Vite installation
npm list vite                     # Check if Vite is installed
npm install vite --save-dev       # Reinstall Vite if needed

🎉 Ready for Jury Evaluation!

Just run docker-compose up --build and visit http://localhost:8080

Complete PDF insights and podcast platform in a single command 🚀

---

🏆 Connecting the Dots - Grand Finale Submission

Delivering advanced AI-powered document processing with modern UI/UX design

📧 Support: Check logs with docker-compose logs -f for any issues