The importance of modeling in computational musicology

Abstract

In recent years, the increasing interest of music theorists in algorithmic analysis as well as the growing amount of musical corpora lead to advances in the field of computational musicology.

I present two examples for the application of mathematical and probabilistic models to musical data. Focusing on notes in Western classical music, I introduce a probabilistic model for pitch-class distributions of musical pieces based on the topological structure of the Tonnetz. The model allows for a quantitative comparison between pieces as well as historical comparisons between different composers. Turning to distributions of chords in 19th-century piano compositions, I demonstrate that they can be modeled using power laws and compare several composers under this model. These examples serve to show how explicit modeling not only aids concrete research questions at hand but also exposes which aspects are not yet fully understood and need to be addressed in future research.

Date
Dec 9, 2020 6:00 PM — 6:20 PM
Location
Universidade Federal do Rio de Janeiro [online]
Rio de Janeiro, Brazil
Avatar
Fabian C. Moss
Postdoctoral Researcher

Fabian C. Moss is a postdoctoral researcher in the Digital and Cognitive Musicology Lab (DCML) at École Polytechnique Fédérale de Lausanne (EPFL, Switzerland). He was born in Cologne, Germany, and studied Mathematics and Educational Studies at University of Cologne, and Music Education (Major Piano) and Musicology at Hochschule für Musik und Tanz, Köln. Working with large symbolic datasets of musical scores and harmonic annotations, he is primarily interested in Computational Music Analysis, Music Theory, Music Cognition, and their mutual relationship.