DocuMind is an AI-based application that allows users to upload PDF documents and ask questions from them. It uses OCR and Retrieval-Augmented Generation (RAG) to provide accurate, context-based answers.
π https://huggingface.co/spaces/AjayArmugam/DocuMind
π https://colab.research.google.com/drive/1JM_KqXbKPf_ola4XUWhlSOR7idW4q4cL?usp=sharing
- π Upload PDF (up to 50 pages)
- βοΈ Extract text from handwritten + printed content (EasyOCR)
- π§ Semantic search using FAISS
- π¬ Chatbot interface for asking questions
- π Source-based answers (no hallucination)
- π Fully deployed and accessible online
- Python
- Gradio
- FAISS
- EasyOCR
- LangChain
- Hugging Face Spaces
DocuMind is an AI-powered document question-answering system that enables users to upload PDFs and interact with them through a chatbot interface.
The system combines OCR (for extracting printed and handwritten text) with Retrieval-Augmented Generation (RAG) using FAISS to perform semantic search and return accurate, context-based answers from the document.