Computational music analysis from first principles

TL;DR

This paper presents a computational approach using coupled hidden Markov models to automatically analyze and annotate a large corpus of Bach chorales, achieving high accuracy without human input and contributing to music theory debates.

Contribution

It introduces a novel automated method for harmonic analysis that minimizes human bias and provides high-accuracy annotations for classical music corpus.

Findings

Achieved over 85% accuracy in chord and key identification.

Provided annotations suitable for music-theoretical analysis.

Demonstrated the effectiveness of minimal-assumption models in music analysis.

Abstract

We use coupled hidden Markov models to automatically annotate the 371 Bach chorales in the Riemenschneider edition, a corpus containing approximately 100,000 notes and 20,000 chords. We give three separate analyses that achieve progressively greater accuracy at the cost of making increasingly strong assumptions about musical syntax. Although our method makes almost no use of human input, we are able to identify both chords and keys with an accuracy of 85% or greater when compared to an expert human analysis, resulting in annotations accurate enough to be used for a range of music-theoretical purposes, while also being free of subjective human judgments. Our work bears on longstanding debates about the objective reality of the structures postulated by standard Western harmonic theory, as well as on specific questions about the nature of Western harmonic syntax.