AI PDF Chatbot

A Streamlit-based PDF question-answering app that uses retrieval-augmented generation (RAG) with ChromaDB and Hugging Face models.

Features

• PDF upload
• ChromaDB vector storage
• RAG-based question answering
• Hugging Face embeddings
• Streamlit chat UI
• Source chunk display for citations
• Conversation memory with st.session_state
• Multi-PDF upload support

Tech Stack

• Python
• Streamlit
• ChromaDB
• LangChain
• Transformers

Setup

1. Create and activate your virtual environment.
2. Install dependencies:

pip install -r requirements.txt

3. Run the app:

streamlit run app.py

Project Structure

docuvoice/
├── app.py
├── ingest.py
├── rag.py
├── document.txt
├── requirements.txt
├── .gitignore
└── chroma_db/

Screenshots

Add screenshots of:

• the PDF upload screen
• the chat interface
• retrieved sources/citations
• multi-PDF workflow