The Annotated Mozart Sonatas: Score, Harmony, and Cadence

This article describes a new expert-labelled dataset featuring harmonic, phrase, and cadence analyses of all piano sonatas by W.A. Mozart. The draws on the DCML standard for harmonic annotation is being published adopting FAIR principles Open Science. annotations have been verified using data triangulation procedure which presented as an alternative approach to handling annotator subjectivity. suited ensuring consistency, within beyond, despite high level analytical detail afforded employed syntax. harmony labels also encode contextual information are therefore investigating music theoretical questions related tonal makeup cadences in classical style. Apart from providing basic statistical characterizing dataset, its potential illustrated two preliminary experiments, one terminal harmonies other relation between performance durations density. Furthermore, particular features can be selected produce more coarse-grained training data, example chord detection algorithms that require less detail. Facilitating dataset’s reusability, it comes with Python script allows researchers easily access various representations tailored their needs.