Tabletalker: Chatbot with Gemini Pro, Streamlit, & LangChain

Inspiration

The root of the TableTalker project design is possibly the intention of creating a chatbot that, apart from being efficient in information retrieval, feels like it has a personality and is talking to you naturally. Through its partnership with the Gemini Pro platform, TableTalker aspires to reduce the divide between digital data and human dialogue by making the transition from digital data to human conversation effortless and compelling, just as talking to a knowledgeable friend.

What it does

answers human queries about database in a humanized manner including memory.

How we built it

Gemini Pro: Google’s advanced language model for natural language understanding. Streamlit: A Python library for creating interactive web applications. LangChain: A Python library for working with language models and embeddings.

Challenges we ran into

training the dataset, text to SQL conversion.

Accomplishments that we're proud of

all the requirements are met as planned. And chatbot can answer about 10-12 tables in a single database, hence including huge memory.

What we learned

data science libraries, use case depth, and environment building.

What's next for Tabletalker: Chatbot with Gemini Pro, Streamlit, & LangChain

This project demonstrates how to build a conversational chatbot using Google Gemini Pro, Streamlit, and LangChain. The chatbot is designed to answer queries related to an SQLite sample database containing multiple tables. It provides human-like responses, making interactions intuitive and user-friendly.

Built With

• google-gemini-pro
• langchain
• python
• sqlite-database
• streamlit
• visual-studio