All our groupmates share a passion for music. Listening on Spotify, practicing instruments, even participating in choirs--music is an integral part to our daily lives. This made us especially curious about what more music could do. What statistics could we find on some of our favorite music? What emotions does a certain artist evoke? Is it possible to count the number of unique words in any given song? These are all questions we kept in mind during this 36-hour hackathon. Using a combination of online databases, machine learning, and web design, this is our Google for music. During this hackathon, we wanted to create a sleek website that is, in essence, Google for music.
What It Does
Upon loading, the search engine gives two routes. Users can either search a song or artist. Analyzing our songs return statistics including the lyrics, word count, # of unique words, the most used words, the average word length, and the longest word present. These were all computed using a variation of machine learning models and Django functionality. The artist page, once again using back-end machine learning, returns the artist's top three tracks based on Internet searches.
How we built it
We built something.
Challenges we ran into
With this year's hackathon being remote, our group faced several situational challenges. Without being in-person, our group had to spend extra time communicating updates, synchronizing our projects and organizing our Github repository. Each member brought unique strengths to the table, which contributed the final project. However, this also imposed a learning curve. We all had to learn Django, set up Pycharm, and find multiple databases within the 36-hour time limit.
Accomplishments that we're proud of
We are proud of the styling of our webpage.
What we learned
For most of us, it was our first time using Django.
What's next for LyricSearch
We want to finish LyricSearch so that it is actually functional. We also want to add an artists page.