ReadNQuiz

ReadNQuiz is a comprehensive Flutter application that transforms how users interact with PDF documents. Users can upload PDFs, generate FAQs, test their knowledge with auto-generated quizzes, and have natural conversations with their documents through our custom RAG (Retrieval-Augmented Generation) system.

✨ Features

User Authentication:

• Secure login and registration powered by Firebase Authentication.

• User profile management and document history.

• Persistent sessions across devices.

PDF Management:

• Upload and store PDFs securely.

• Organize documents in personalized collections.

Smart FAQ Generation:

• Automatically extract and generate frequently asked questions from uploaded PDFs.

• Key concept identification for study focus.

• Export FAQs for offline reference.

AI-Powered Quiz Generator:

• Create custom quizzes from any uploaded PDF.

• Multiple question formats (multiple choice).

• Quiz performance tracking and analytics.

Conversational PDF Interface (RAG System):

• Ask natural language questions about your PDF content.

• Get precise answers with relevant context.

• Support for complex, multi-part questions.

• Smart response handling when information isn't in the document.

🛠️ Tech Stack

Frontend: – Flutter for cross-platform mobile application.

Backend: –

• FastAPI for the RAG system and quiz generation.

• Firebase for authentication and document storage.

AI/ML: –

Google Gemini for context-aware quiz generation

Custom RAG implementation using:

• FAISS for vector similarity search.

• HuggingFace embeddings (all-MiniLM-L6-v2).

• Text chunk processing and retrieval optimization.

🚀 Getting Started

Prerequisites

Flutter SDK (version 3.0+)

Firebase account

Google Gemini API key

Python 3.8+ (for backend services)

Installing

Clone the repository to your local machine:

  git clone https://github.com/shivansh00011/ReadNQuiz.git
  cd ReadNQuiz  

Install dependencies: Run the following command in your project directory to fetch all required dependencies:

flutter pub get

Set up the backend: You can take the backend code from my repository named "ReadNQuiz-Backend".

Configure environment variables:

• Create a .env file in the backend directory

• Add your Gemini API key: GOOGLE_API_KEY=your_key_here

• Add your Firebase configuration settings

Run the backend server:

uvicorn main:app --reload

Launch the Flutter application

Architecture Overview

Mobile Application: Flutter-based application.

Cloud Backend: FastAPI server hosting our RAG system and quiz generation logic.

Firebase Services: Authentication, storage, and realtime database for synchronization.

How does the RAG works:

The RAG system works in two phases:

Indexing Phase: When a PDF is uploaded, it's processed, chunked, embedded, and stored in a vector database.

Query Phase: When a question is asked, it retrieves relevant chunks from the vector database, constructs context, and generates an answer using Google Gemini.

Project Video:

ReadNQuiz.mp4

🤝 Contributions

We welcome contributions! If you'd like to improve the project, feel free to fork the repository and submit a pull request.