Disclaimer: We're still talking with officials about weather or not this project should be judged. We're getting conflicting answers.


We wanted our professors to read our textbooks to us, like audio books. Who wouldn't!?

What it does

Given a YouTube link and a paragraph to read, squidsplit will analyze the words in the video and stitch them together in the order you choose!

How We built it

We used Pydub for most of the audio manipulation and IBM Watson to analyze the speech as text. We used Flask for our backend and our complementary group used vue.js to make a nice frontend. Our backend can be used fully independently from the front end, and we could make a command line interface or a Discord bot relatively easily.

Challenges We ran into

The IBM Watson API was difficult to use because there wasn't enough documentation. It was also difficult to get back into Python after not using it for several months.

Accomplishments that We're proud of

We're so glad we were able to get our project into a fully working state by the deadline. There were many delays, but we stayed up overnight and were able go get it working! It's so awesome that it can analyze speech and stitch things together. It doesn't sound very realistic, but that's not what we were going for. I'm glad we were able to get some entertainment out of our project.

What We learned

We learned how to use the Watson API and manipulate audio files.

What's next for squidsplit

We'd like to turn it into a Discord bot so it's easier for everyone to access it quickly.

Disclaimer: We worked with another team together within one repo. We focused on the backend, while the other team focused on the front end. Our entry for this project is for the backend specifically. We'd like to thank them specially; without them, we wouldn't have had such a user friendly website, so we'd like to credit them here. Special thanks to Rin Krivokrysenko and Lana Boyko for working on the front end! Austin Hockenberry also worked with us on the back end, but he's asleep after working so hard. Before we started, we were explicitly told by event coordinators that it was okay to split our group like this. After hearing the live stream during Since there was 6 of us working in the same repo we do not think that we should be eligible for awards, this was a project we thought would be fun and a good experience to make.

Built With

Share this project: