Inspiration
This project was inspired by being behind in some of our classes. Although reading transcripts can be helpful, we thought providing succinct summaries of important information would be usefull.
What it does
Our web app allows users to input any text into a text box, then submit it to the OpenAI GPT-3 alogirthm for processing. We split the text into smaller elements so the results are more accurate and the result is generated faster. We ran these elements asynchronously to allow for the model to execute concurrently.
How we built it
We built this app with Flask. Other technologies included Tailwind CSS, Python, and HTML.
Challenges we ran into
We ran into a lot of challenges configuring Tailwind to work with Flask. It was also hard to tune the model to get the best results. It required a lot of iterations to get the correct result.
Accomplishments that we're proud of
We are proud of challenging ourselves to build a complete web app that involved multiple elements. We had always heard of the cool algorithms created by Open AI but never gotten the chance to put them into practice. This project was a great opportunity for us to explore the use cases of GPT-3.
What we learned
This project challenges us in multiple ways, and we learned a lot about the different parts required to create a complete web app. We experienced some very frustrating bugs along the way but learned that talking out the problem to each other yieled the best results.
What's next for Quick Lecture
We believe that this app and concept have a lot of potential. With the amount of content being produced these days, there is an incredible opportunity for us to improve how we read and make use of that information. Expanind on this app would include improving the UX and fine tuning the model.
Log in or sign up for Devpost to join the conversation.