Inspiration!!
We are 3 MIT students who want to learn as quickly and effectively as possible in today's fast-paced atmosphere. We all love OCW and, on more than one occasion, have had to use x2 or even x4 playback speed to process information faster. However, we also noticed that faster playback speed comes with the risk of reduced information intake. We seek a solution that is more nuanced than just a speed multiplier.
What it does?
Our tool cuts out the white space during lecture videos in such a such a way so that all the information is condensed, but not sped up, to make the most out of the short time we have in life. We remove moments in lectures where professors are not speaking and simply writing on the chalkboard or just walking across the room.
How we built it?
We used RevSpeech API to determine the transcript of the YouTube video. In the transcript, we then have the times for the beginning and ends of each word. Using this information, we can determine the white space between the words and determine if it is too long and needs to be shortened. The back end is developed in Python, and we are hosting it on an Azure web app (hopefully). It works offline and we will strive to host it online.
Challenges we ran into
Being able to develop a front end and a back-end that works!!!
Accomplishments that we're proud of
Successfully creating an offline (possibly online) solution for this problem!!!!
What we learned
Being a developer takes patience and time. And lack of sleep. And coffee. We're still learning !! :D
What's next for Videdu3
We will optimize the algorithm for the back end to create a more efficient system that not only takes silent parts of lectures into consideration but also the time it takes to pronounce words. That way we can optimize the learning process even more without the drawbacks of the standard solution of a speed multiplier.
Log in or sign up for Devpost to join the conversation.