Inspiration
University students are busy; they simply do not have time to write notes by hand all the time. Furthermore, students depend on YouTube more often than ever, as learning on-the-go seems to be the more appreciated method of learning than listening to a professor lecture all day. That's why we've created a revolutionary new application that generates a set of notes for you out of a YouTube video!
What it does
This transcriber implements functionality that scans the URL of a desired YouTube video and parses the file of timed text obtained from the video. The information is parsed, and the captions themselves are stored into an array of single sentences. The application also implements additional functionality that allows the user to obtain a summary of a YouTube video by providing any number of keywords. The program will then search every single sentence of the audio transcript that contains a significant number of occurrences of each keyword and of its synonyms. Each of these relevant sentences are then aggregated into a single text file in which they will be displayed as bullet points. The user can treat the bullet points as a textual synopsis of the YouTube video.
How we built it
We used Python to code our entire project. Various IDEs that we used include IDLE, Visual Studio Code, and Eclipse. We used pip to install third-party libraries pytube, which contains functionality to connect to YouTube and obtain information on YouTube videos, and ntlk, which contains a file that loads the definitions, synonyms, and antonyms of words.
Challenges we ran into
We originally wanted to utilize Amazon Web Services (AWS) to include artificial intelligence in voice recognition and language translation in our application. However, AWS consistently denied us permission for us to use their cloud services, so instead, we stuck with offline Python tools.
Log in or sign up for Devpost to join the conversation.