Song selection is essential to any karaoke party. To help you, Daniel Knopf ’22, Nalu Tripician ’22 and Isabel Wrubel ’22 have recently redesigned and launched an app, Versatilitywhich organizes a playlist of the top 10 karaoke songs from a user’s Spotify library.
To create playlists for users, the app’s algorithm rates songs based on their “karaoke suitability,” meaning their similarity to tracks known to be good for karaoke. It compares them using Spotify’s publicly available metrics, including tempo, key, and “danceability.”
“We’re using a neural network that we’ve designed that basically takes everything [those metrics]…then spits out a karaoke score,” Tripician said.
After kicking off the project at a hackathon in 2019, Tripician and Knopf worked on the app on and off until summer 2021.
Each member of the team brought something different. For example, Tripician mainly worked on the app’s algorithm.
“My role was more to work on the back end,” Tripician said. “I [developed] the machine learning algorithms that say whether [a song is] a good karaoke song or not, and also I take care of everything on the Spotify side, like creating playlists. Then I pass… the data I receive to Daniel’s side, so he can display it.
Knopf works on web development, which involves storing users’ music data on a server and displaying their results on the app’s website.
“The page that users visit… is verseatility.co, and that’s the front-end web design, which I built in this framework called React,” Knopf said. “Then we have a main server, which is separate. We put a running computer somewhere else that has all our data…. I connect them so that all data is shared between the two in a safe and secure manner. »
Wrubel joined the team in the summer of 2021 to revamp the site’s graphics. Jhe apps UI has been kept minimal, with a light green radial gradient to give a karaoke feel to their website and mimic the Spotify Wrapped experience.
“I’m largely responsible for the design,” Wrubel said. “I’m working in Figma to build a design using [user experience and user interface] principles.”
Wrubel needed to keep Tripician’s work with machine learning in mind when designing the website.
“So that our [machine learning] algorithm to improve, we needed users to rate each individual song,” Wrubel wrote in an email to The Argus. “It was delicate of a [user experience] perspective. [I] decided to make the stream a ten-page series where a user could only get their entire playlist once they individually rated each of their songs.
Initially, the team calibrated the algorithm using a web scraper, a software tool that harvests data from websites, to find good karaoke songs. They also manually rated over 1,000 songs, which Tripician put together. Some of these tracks, including classical music, were quickly given a low rating.
“[Tripician] put the whole Minecraft soundtrack in [to be rated]”, Knopf said. “I was like, ‘All right, automatic zero.'”
The app also asks users to rate the songs they receive, and the algorithm adjusts its recommendations accordingly.
The team noted that some users were getting lackluster playlists because they just didn’t listen to a lot of karaoke music. Another problem that has emerged is that users sometimes do not remember their Spotify username and password, which they have to enter on the website.
Overall, however, the team hopes gaining users will help them uncover lingering bugs in the website’s design.
“What we really want to hear is if you use it and it doesn’t work,” Knopf said. “We would love for you to contact us. We just added some data logging and analysis…so we can try to figure it out.
Going forward, the developers want to expand the scope and capabilities of the app.
“We’re working on adding a few more features like getting an overall karaoke score for your entire library,” Knopf said.
Other goals include expanding playlists to 25 or 50 tracks, allowing users to view karaoke ratings for specific songs, and displaying aggregate song and artist data on the website. The team is considering a model similar to Spotify Wrapped, where users can post a list of their top karaoke songs to social media.
“We think it…has a lot of potential to be shared and used by people outside of Wesleyan,” Knopf said. “We’re trying to add more features and figure out what’s the best way to make it big.”
Nalu Tripician is the web editor of The Argus.
Daniel Knopf was an associate opinion editor for The Argus.
Caleb Henning can be reached at [email protected]
Anne Kiely can be reached at [email protected]