Media, Pennsylvania – Will accurate timing and inaccurate emotions make better music? Artificial intelligence (AI) may be able to help answer questions, according to Michael Yatauro, an associate professor of mathematics at Pennsylvania State Brandywine, and Arnav Pandey, a mathematics major in his second year transitioning from Brandywine to University Park.
The two are collaborating to see how drummers' performances compare to accurate metronome times and were selected to present their work at the American Mathematics Association's Joint Mathematics Conference in Washington, DC in January.
“Alnab and I are both drummers. He has experience in marching bands and is in a blue band, but I have a rock band background,” Yatauro said. “We're going to land on this project, inspired by a paper that explores how music goes about emotions rather than accurate timing. It's not accurate, but not complete, not accurate beats. We're looking to explore ways to analyze statistically different beats.
Yatauro and Pandey used an AI tool called Moises, a toolkit for musicians who take the soundtrack and destroy it with instruments. With this tool, I was able to only listen to the drum beats of the song.
“The good thing about drumming is that it's very staccato in terms of sound. When the drum is hit, it creates a waveform that can be analyzed more accurately,” Yatauro said, explaining that it uses moises to analyze the waveform. “I don't want to hear or pause because I'm doing this for a full drum set, as it's going to create a lot of inconsistencies and errors. Instead, I'll use Moise to isolate the drum tracks and feed the waveforms to Google Colab.
They still plan how the findings from this study will be used. Currently, they are running tests on sample data (song clips) to test the limits of the process and identify areas of improvement or adjustment.
From an application perspective, Yatauro said that using this study could produce more realistic sound drum machines due to the style of music that uses them.
“We are also looking to understand the types of timing that produce music that is perceived as 'attractive'. This will be the process of studying randomness in the context of music,” he said. “Some people call this randomness “pink noise” and distinguish it from other forms of randomness, such as white or brown noise. ”
Pandee is studying at University Park, so he and Yatauro meet once a week on Zoom.
“The research project is extremely fulfilling for me as I was able to prove to myself that mathematics can be applied to creative outlets,” Pandi said. “Two huge fields of my life, music and scholars have always been my goals. Dr. Yatauro and I are both drummers who love mathematics, so this project has been pretty much waiting for me.”
Pandi also noted that the project provided education beyond classrooms.
“Learning about the research process, collecting data from different music charts manually, and trying out different different methodologies has provided an invaluable learning experience, unlike anything we've gained from traditional classes,” Pandi said.
