AI Study Buddy is an AI-powered educational assistant that helps students study more efficiently from lecture materials. It uses Retrieval-Augmented Generation (RAG) to answer questions accurately based on uploaded PDF notes rather than relying only on the language model's general knowledge.
The application allows users to upload lecture PDFs, ask questions about the content, generate concise summaries, create AI-generated multiple-choice quizzes, generate flashcards for revision, build personalized study plans, and summarize educational YouTube videos using their transcripts.
To provide accurate answers, the system extracts text from PDF files, divides it into overlapping chunks, converts each chunk into semantic embeddings using Sentence Transformers, and stores these embeddings in a FAISS vector database. When the user asks a question, the system retrieves the most relevant chunks and sends them, along with the question, to the Mistral Nemo language model to generate a context-aware response.
The project is developed in Python with a Streamlit interface and integrates several AI technologies, including Hugging Face Transformers, FAISS, Sentence Transformers, PyTorch, and PyPDF2, to provide an interactive and intelligent learning experience.