Year: 2018
Type: article
Abstract: 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.
Source: Journal of New Music Research
Authors: Hiroaki Tsushima, Eita Nakamura, Katsutoshi Itoyama, Kazuyoshi Yoshii
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: 音楽文法と楽譜変換の階層ベイズモデルに基づく編曲技能の計算論的解明, Proposal of Representation Method for Time-Series Media Based on Tree Structure and Realization of its Calculus, A Computational Model of Music Understanding Based on Statistical Grammar and Constructive Semantics, Expansion of "PEDB" a Music performance database, Statistical Music Modelling and Automation Techniques for Expressive Performance and Accompaniment Generation Capturing Performers' Musicality, G432938063, G5300062852, Emergence of Musical Intelligence Based on Probabilistic Integration of Signal and Symbolic Processing, Research and development of practice system for piano beginners using vertical line notation
OpenAlex ID: W2745229848