Abstract by Brooks Butler

Personal Infomation

Presenter's Name

Brooks Butler


Eric Todd
Taylor Kimball

Degree Level



Eric Todd
Taylor Kimball

Abstract Infomation


Physics and Astronomy

Faculty Advisor

Kent Gee
Mark Transtrum
Sean Warnick


Unsupervised classification of crowd noise at BYU basketball games


The relationship between crowd noise and crowd behavioral dynamics is a relatively unexplored field of research. Signal processing and machine learning (ML) may be useful in classifying and predicting crowd emotional state. As a precursor to performing ML, it is instructive to identify which crowd acoustic events an unsupervised ML algorithm would classify as unique. Audio features have been extracted from recordings of crowds at Brigham Young University basketball games and data collection methods are discussed. A k-means clustering analysis was conducted on half-second segments of the recordings using these extracted audio features. For example, a clustering analysis performed on a one-twelfth-octave spectrogram of crowd noise recordings reveals there are approximately six unique events that occur during a game. Implications for further ML algorithm development based on various sets of audio features are discussed.