Clairify

Inspiration

In our experience, reading 20+ page documents at school or for research has always been a daunting task. Even after thoroughly reading the document, most of the important information isn't retained. We wanted to create a chatbot to help better digest the information and answer specific questions about the document, website, or video.

What it does

Once you upload a document, website URL, or YouTube video, our Streamlit app allows you to ask any specific questions about the topic to the chatbot. It provides information only from the uploaded information and lists the specific document source.

How we built it

We built the chatbot using Langchain, Pinecone, and Openai libraries in Python. The app is built using Streamlit.

Challenges we ran into

We ran into challenges with using OpenAI keys, incorporating an option for YouTube videos, streamlit, and ensuring document specificity.

Accomplishments that we're proud of

We're proud of incorporating multiple media options and creating a user-friendly UI.

What's next for App

In the future, we will look to include information on where the answer was extracted from the lesson and highlight the specific lines.

Built With

• langchain
• openai
• pinecone
• python
• streamlit