AI-Powered Document Summarizer Development

Bindupautra Jyotibrat

Completed work

AI Model Developer

AI Application Developer

Web Developer

LangChain

React

spaCy

Artificial Intelligence

📄 Document Summarizer

AI-powered document summarization tool that extracts key insights from PDFs, DOCX, and plain text files.

📑 Table of Contents

🚀 Installation & Setup

🎨 Figma Design

🎥 How to Run the Application

🌟 Features

🔹 Key Highlights

✅ Extract text from PDFs & DOCX files effortlessly. ✅ AI-generated summaries using state-of-the-art transformer models. ✅ Intelligent keyword extraction for enhanced insights. ✅ Multi-model support (T5, BART, Pegasus, etc.). ✅ Vector search for precise retrieval. ✅ Sleek React.js frontend for a seamless experience. ✅ Question Answering support.

🏗️ Tech Stack

🔹 Backend

Flask - Lightweight web framework

Transformers (Hugging Face) - Pretrained NLP models (T5, BART, Pegasus)

PDFPlumber - Extracts text from PDFs

docx - Parses DOCX files

spaCy & YAKE - NLP processing & keyword extraction

LangChain & FAISS - AI-powered document retrieval & vector similarity search

Sentence Transformers - Embedding generation

Gemini 1.5 Flash API - Used for Question Answering

🔹 Frontend

React.js + Vite - Blazing fast UI development

Tailwind CSS - Modern styling framework

Lucide Icons - Minimalist UI enhancements

📂 Project Structure

Document-Summarizer/
│
├───.gitattributes
├───.gitignore
├───eslint.config.js
├───index.html
├───LICENSE
├───package-lock.json
├───package.json
├───postcss.config.js
├───tailwind.config.js
├───tsconfig.app.json
├───tsconfig.json
├───tsconfig.node.json
├───vite.config.ts
│
├───.github/
│    └───README.md
│
├───Archive/
│   ├───Document_Summarizer_first_prototype.ipynb
│   ├───Document_Summarizer_second_prototype.ipynb
│   ├───summarizer_web.ipynb
│   └───README.md
│
├───Assets/
│   ├───Contributors/
│   │   ├───Akshit Github Photo.png
│   │   ├───Ansh Github Photo.png
│   │   ├───Bindupautra Github Photo.png
│   │   └───Rana Github Photo.png
│   │
│   ├───Videos/
│   │   ├───Demo.mp4
│   │   └───To Run in Local Machine.mp4
│   │
│   └───README.md
│
├───backend/
│   ├───.env
│   ├───app.py
│   ├───Readme.md
│   ├───requirements.txt
│   └───README.md
│
├───NoteBooks/
│   ├───Summarizer.ipynb
│   └───README.md
│
└───src/
    ├───App.tsx
    ├───index.css
    ├───main.tsx
    ├───vite-env.d.ts
    └───README.md

🚀 Deployment Status

🔹 Frontend Deployment

The frontend of the application is deployed and accessible for preview. You can explore the user interface and interact with the design at the following link:

👉 Live Frontend Preview

However, please note that the backend has not been deployed yet. As a result, the full functionality of the Document Summarizer, including AI-powered summarization and document processing, is not available in the live version.

To experience the complete features, you will need to set up the backend locally by following the Installation & Setup instructions.

🚀 Installation & Setup

🔧 1️⃣ Clone the Repository

 git clone https://github.com/Jyotibrat/Document-Summarizer.git
 cd document-summarizer

🔑 2️⃣ Obtain a Gemini API Key

To use the Gemini 1.5 Flash API for question answering, you need to obtain an API key. Follow these steps:

Sign Up for Gemini API:

Go to the Gemini API website and sign up for an account.

Follow the instructions to create a new API key.

Set Up the API Key:

Navigate to the backend directory:

cd backend

Open the .env file in a text editor:

Add your Gemini API key to the .env file:

GEMINI_API_KEY='your_api_key_here'

Save and close the file.

🖥️ 2️⃣ Backend Setup

Navigate to the backend directory.

cd backend

Create a virtual environment and install dependencies:

 python -m venv venv
 source venv/bin/activate
 pip install -r requirements.txt

Run the Flask server:

 python app.py

🎨 3️⃣ Frontend Setup

Navigate to the root directory and install dependencies:

 npm install

Run the development server:

 npm run dev

🎨 Figma Design

The UI/UX design for the Document Summarizer was created using Figma. You can view the design and interact with the prototype using the link below:

👉 Figma Design