PDF-Insight: 📑 AI PDF Highlight Summarizer

DEMO LINK: https://pdf-insight.streamlit.app/

🚀 Overview

This project extracts highlighted text from a PDF and generates a concise summary using Mistral-7B from Hugging Face. Built with Streamlit, it provides an easy-to-use interface for document analysis.

🔥 Features

• 📂 Upload PDF and extract only the highlighted text
• 🤖 Summarize extracted highlights using Mistral-7B
• 🔑 Requires Hugging Face API key for Mistral-7B access
• ⚡ Multi-user support using st.session_state
• 🎨 Streamlit UI for an interactive experience

🛠 Installation

Using Poetry

# Clone the repository
git clone https://github.com/ShubhamMandowara/PDF-Insight.git
cd PDF-Insight

# Install dependencies using Poetry
poetry install

# Activate virtual environment
poetry shell

# Run the Streamlit app
streamlit run app.py

Using pip

pip install -r requirements.txt
streamlit run app.py

📝 Usage

1. Enter your Hugging Face API Key in the sidebar
2. Upload a PDF containing highlighted text
3. View extracted highlights
4. Click "Summarize Highlights" to generate an AI-powered summary

📦 Requirements

• Python 3.8+
• Streamlit
• PyMuPDF (pymupdf)
• LangChain
• Hugging Face Hub
• langchain_community

📜 Example Output

**Extracted Highlights:**
1. "Artificial Intelligence is transforming industries."
2. "Mistral-7B is a powerful open-weight LLM."

**Generated Summary:**
AI is revolutionizing various sectors, and Mistral-7B is a strong model in the field.