Inspiration
We took our inspiration from students of various ages who struggle with understanding topics and reading the book. We wanted to make a pedagogical tool that can curtail lessons to a user by taking into account of their learning needs.
What it does
We implemented a program that helps curate to students needs where they are able to upload a PDF file and the program will read back the file. Where it can then take questions from the asker and provide accurate answers from the textbook files in video form
How we built it
Our application uses the streamlit python library that helped us develop the framework for producing images, and videos. Our implementation of MindsDB helped with the transition of text to image. We use LangChain and the vector database Pinecone to store documents efficiently. We then query the database for a specific book and a corresponding question with regards to the content of the book. The result we get from the query gets passed back to LangChain and then it creates the script for the text to speech and the image prompts for both the image scraper and the text to image ai that is hosted in MindsDB. All of these elements are combined, and the result is a video lesson generator.
Challenges we ran into
Most of the challenges have been getting Poetry installation to work so the team can all have the same versions of the files. We had a lot of issues with combining the front-end and the back-end. In the end, we weren't able to consolidate the two completely.
Accomplishments that we're proud of
The success we are proud of is getting the program to read the file and able to generate AI voice to read out the information along with the image. We are also proud about utilizing new frameworks like streamlit, Pinecone, MindsDB, and LangChain. It was difficult to combine all of these complicated components together in order to make a coherent hole. But we are proud that we were able to combine all of these techs to build our primary function successfully.
What we learned
There are a lot more implementations for us to add to the application as it has big potential in the academic setting in helping students learn.
What's next for Kon Academy
We hope to plan to make this with more features and also transition into a mobile App where it's easily accessible on the go. We also had implementations to create flashcards and quizzes, but we did not have enough time to implement them visually.
Log in or sign up for Devpost to join the conversation.