📄 PDF AI Summarizer

A powerful Python application that uses OpenRouter (access to 300+ LLMs via an OpenAI-compatible API) to automatically summarize PDF documents. Upload any PDF and get intelligent, customizable summaries in multiple formats!

✨ What This Project Does

PDF AI Summarizer is an intelligent tool that:

• Reads PDF Files: Extracts all text from PDF documents automatically
• Cleans & Processes Text: Removes unnecessary characters and formats text for analysis
• Splits into Chunks: Breaks large documents into manageable pieces for the AI
• Generates Smart Summaries: Uses OpenRouter (OpenAI-compatible API) to create summaries in different styles:
  • Brief: Quick 2-3 sentence overview
  • Detailed: Comprehensive summary with key points, arguments, and conclusions
  • Bullet Points: Organized, easy-to-scan summary format
  • Executive: Business-focused summary with insights and recommendations
• Analyzes Document Structure: Identifies document type, main topics, and target audience
• Extracts Key Quotes: Pulls out important statements and phrases from the document
• Exports Results: Download summaries as text files or statistics as CSV

Perfect for students, professionals, researchers, and anyone who needs to quickly understand the content of large PDF documents!

---

🚀 Getting Started

Prerequisites

Before you start, make sure you have:

• Python 3.8 or higher installed on your computer
• An OpenRouter API Key (get one at openrouter.ai/keys)

Step 1: Clone the Repository

git clone https://github.com/niti007/pdfsummarizer.git
cd pdfsummarizer

Step 2: Create a Virtual Environment (Optional but Recommended)

This keeps your project dependencies isolated:

On macOS/Linux:

python3 -m venv venv
source venv/bin/activate

On Windows:

python -m venv venv
venv\Scripts\activate

Step 3: Install Dependencies

Install all required packages:

pip install -r requirements.txt

What gets installed:

• streamlit - Interactive web interface
• langchain & langchain-openai - AI framework
• PyPDF2 - PDF text extraction
• python-dotenv - Environment variable management
• pandas - Data handling
• tiktoken - Token counting

Step 4: Set Up Your OpenRouter API Key

1. Go to OpenRouter Keys
2. Click "Create Key"
3. Copy your API key

Create a .env file in the project root directory:

echo "OPENROUTER_API_KEY=your_api_key_here" > .env

Replace your_api_key_here with your actual API key.

⚠️ Important: Never share or commit your .env file with your API key!

Step 5: Run the Application

Start the Streamlit web application:

streamlit run app.py

This will open a browser window at http://localhost:8501 where you can use the application.

---

📖 How to Use

Using the Web Interface

1. Upload a PDF: Click "Choose a PDF file" and select your document
2. Configure Settings (optional):
   • Summary Type: Choose from brief, detailed, bullet points, or executive summaries
   • Max Tokens per Chunk: Adjust processing speed and depth (default: 8000)
   • Show Document Analysis: Enable to analyze document structure
   • Extract Key Quotes: Enable to extract important quotes
3. Process: Click "🚀 Process PDF" button
4. View Results: See the summary, analysis, and statistics
5. Export: Download results as text or CSV file

Example Workflow

1. Upload a 20-page research paper
2. Select "Executive" summary type
3. Click Process
4. Read the AI-generated summary in 2 minutes instead of 20 minutes
5. Download the summary for sharing or reference

---

🛠️ Project Structure

pdfsummarizer/
├── app.py                 # Main Streamlit web application
├── pdf_processor.py       # Handles PDF reading and text cleaning
├── summarizer.py          # AI summarization logic using OpenRouter
├── utils.py              # Helper functions (validation, formatting, export)
├── requirements.txt      # Python dependencies
├── .env.example          # Example environment file
└── README.md             # This file

Key Files Explained

• app.py: The main web interface. Handles file upload, displays UI, and coordinates the workflow
• pdf_processor.py: Reads PDFs, extracts text, cleans it, and splits it into chunks
• summarizer.py: Uses the OpenRouter API to generate different types of summaries
• utils.py: Helper functions for API validation, time estimation, and result formatting

---

💻 Technical Details

How It Works

User uploads PDF
     ↓
Extract text from all pages
     ↓
Clean and normalize text
     ↓
Split into manageable chunks (respecting token limits)
     ↓
Send each chunk to OpenRouter
     ↓
Get individual summaries
     ↓
Combine summaries into one cohesive final summary
     ↓
Display results and export options

Token Management

The app uses tiktoken to count tokens before sending text to the AI. This ensures:

• Text stays within API limits
• Processing is efficient
• Accurate time estimation

---

⚙️ Configuration Options

Summary Types Explained

Type	Best For	Length
Brief	Quick overviews	2-3 sentences
Detailed	Complete understanding	Full coverage of topics
Bullet Points	Quick scanning	Organized key points
Executive	Business decisions	Key insights & recommendations

Max Tokens Setting

• Lower (4000-5000): Faster, less detail
• Default (8000): Balanced speed and quality
• Higher (9000-10000): More detail, slower processing

---

🔐 Security & Privacy

• Your API key is stored locally in .env (never committed to git)
• PDFs are processed locally; nothing is permanently stored
• Summaries are generated server-side by OpenRouter

---

🐛 Troubleshooting

"OPENROUTER_API_KEY not found"

• Make sure .env file exists in the project root
• Verify the key is correctly set: OPENROUTER_API_KEY=your_actual_key
• Don't use quotes around the key in .env

"Error extracting text from PDF"

• Ensure the PDF is not corrupted
• Try opening the PDF in Adobe Reader to verify it's valid
• Some scanned PDFs (images) may not have extractable text

"Rate limit exceeded"

• Google API has daily limits on free tier
• Wait a few hours and try again, or upgrade your API plan
• Consider reducing the number of chunks

Application runs but interface doesn't load

• Make sure port 8501 is available
• Try: streamlit run app.py --server.port 8502

---

📦 Requirements

All dependencies are listed in requirements.txt:

streamlit>=1.28.0
langchain>=0.0.350
langchain-google-genai>=0.0.5
python-dotenv>=1.0.0
PyPDF2>=3.0.1
pandas>=2.0.0
tiktoken>=0.5.0

---

🤝 Contributing

Found a bug or want to suggest a feature? Feel free to:

1. Create an issue describing the problem
2. Fork the repository and submit a pull request

---

📄 License

This project is open source and available under the MIT License.

---

🎯 Future Enhancements

• Support for more document formats (DOCX, TXT, etc.)
• Multiple language support
• Summary comparison between different AI models
• Batch processing for multiple PDFs
• Custom summary templates
• Cloud storage integration

---

📞 Support

For issues or questions:

• Check the Troubleshooting section above
• Open an issue on GitHub
• Review the code comments for technical details

---

🌟 Acknowledgments

• OpenRouter for powerful AI summarization
• Streamlit for the beautiful web interface
• LangChain for AI orchestration
• PyPDF2 for PDF processing

---

Happy summarizing! 📚