A conversational chatbot built with Gradio and Hugging Face Transformers that supports multilingual question answering (English, Turkish, French, etc.).
Users can upload PDF, TXT, or DOCX documents and ask questions based on the content.
The chatbot also maintains conversation history for multi-turn dialogues.

• Multilingual question answering using mrm8488/bert-multi-cased-finetuned-xquadv1 model
• File upload support for .pdf, .txt, and .docx formats
• Persistent chat history to remember previous conversation turns
• Built with Gradio for an interactive web interface
• Supports multiple languages (English, Turkish, French, and more)

You can try the chatbot live on Hugging Face Spaces:

Hugging Face Space: multilingual-qa-chatbot

1. Clone the repository:

   git clone https://github.com/ozlemelo/multilingual-qa-chatbot.git
   cd multilingual-qa-chatbot

2. Create and activate a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate # Linux/Mac
   venv\Scripts\activate # Windows

3. Install the dependencies:

   pip install -r requirements.txt

Run the chatbot locally with:

Then open the displayed local URL in your browser to start interacting.

• Upload a .pdf, .txt, or .docx file containing the context you want to ask questions about.
• Enter your question in the input box.
• Press Submit and the chatbot will reply based on the uploaded document.
• Chat history is maintained during the session.

Here is a screenshot of the chatbot in action:

• transformers
• gradio
• PyPDF2
• python-docx

• Switch to a more powerful model like distilbert, mbart, or OpenAI-compatible LLMs
• Add more advanced multi-turn memory (e.g., using vector stores)
• Add support for audio input/output (speech-to-text and text-to-speech)
• Dockerize for production use
• Deploy on other platforms like Streamlit, FastAPI, or Flask

Feel free to open issues or pull requests to improve the project.

This project is licensed under the MIT License.

Ozlem Ekici