Rhythms in Ragtime – investigating genre characteristics with a computational approach to metric hierarchies
Music theory and cognition studies have investigated extensively how musical rhythms induce hierarchies of metric accents. Computational approaches to meter induction in music allow the study of these processes on a large scale. In my talk I present ongoing work with the computational model of Inner Metric Analysis, which assigns a metric weight to each note of a musical piece. The model’s potential to extract meaningful metric hierarchies from music theoretical and cognitive perspectives has been revealed by applying the model to diverse musical pieces (such as Western classical music, e.g. Brahms, Webern, Stravinsky, Haydn, Schubert, Schumann; Latin American dances, Dutch folk songs), and to results of listening experiments. At present, we are studying rhythmic-metric characteristics of Ragtime by applying the model to a large Ragtime corpus of 11.000 MIDI files in the context of Music Information Retrieval. With this research we seek to contribute to the lively discussion of Ragtime fans and scholars on the question “What constitutes Ragtime?” Our computational investigation of this genre, which is considered a “hybrid of folk and written cultures”, contributes to fundamental questions of musical rhythm, such as “what is syncopation”; it also contributes to questions on how to investigate music meaningfully through a “big data” approach within Music Information Retrieval. I will discuss challenges on interconnecting theoretical, cognitive and computational approaches to rhythm research.
presentation slides available here