Generative Statistical Models with Self-Emergent Grammar of Chord Sequences
This work addresses the need for chord categories in harmony models from an informatics perspective, offering an incremental improvement in generative statistical modeling for music processing.
The paper tackled the problem of modeling chord sequences in music by using hidden Markov models and probabilistic context-free grammars with latent variables to capture syntactic similarities among chords, finding that these models often outperform conventional Markov models in predictive power and that the self-emergent categories align with traditional harmonic functions.
Generative statistical models of chord sequences play crucial roles in music processing. To capture syntactic similarities among certain chords (e.g. in C major key, between G and G7 and between F and Dm), we study hidden Markov models and probabilistic context-free grammar models with latent variables describing syntactic categories of chord symbols and their unsupervised learning techniques for inducing the latent grammar from data. Surprisingly, we find that these models often outperform conventional Markov models in predictive power, and the self-emergent categories often correspond to traditional harmonic functions. This implies the need for chord categories in harmony models from the informatics perspective.