Human audition has largely been studied from a neurological and psychoacoustic lens, but with the advent of powerful computational tools, human audition and other sensory systems in the brain can now be studied via deep learning systems. Neural networks in particular have been shown to be successful in predicting brain activity and neural responses to visual stimuli but are less studied in the context of audition. Thus, the overall aim of my project is to leverage machine learning tools trained on human tasks to see whether we can build intelligent systems that replicate human audition and if there are biological insights to be gleaned from these models.

“I am participating in SuperUROP so I can devote myself to a long-term research project as an undergraduate student, specifically at the intersection of applied ML and neuroscience. Prior research experience and classes I have taken have both prepared me for this opportunity, and my project feeds into my interests in music, mathematics, machine learning, and beyond. I’m excited to learn more about the field and see what results my project may yield!”