A Demo application demonstrating question answering over whatsapp conversations using large language models.
pip install -r requirements.txt
Sign up for an OpenAI developer account here. Follow the instructions on the site to get your API key. Then, in your shell, run:
export OPENAI_API_KEY=<your key>
Export a WhatsApp conversation as a text file and save it in the data directory. The file should be named whatsapp_export.txt.
streamlit run streamlit_app.py
The app has two tabs. On the first tab called BasicQA, you will have to provide the snippet of conversation yourself and ask a question about it. There is a character limit of 10,000 characters that I have imposed to keep the request within the 4096 token limit.
On the second tab called QA with memory, the file whatsapp_export.txt is read and the entire conversation is processed into embedding and stored in an in-memory index as well as persisted to local disk under a folder called vector_index. You can ask questions about the conversation.
