Perception-Inspired Graph Convolution for Music Understanding Tasks

Emmanouil Karystinaios,Francesco Foscarin,Gerhard Widmer

2024-05-15

Abstract:We propose a new graph convolutional block, called MusGConv, specifically designed for the efficient processing of musical score data and motivated by general perceptual principles. It focuses on two fundamental dimensions of music, pitch and rhythm, and considers both relative and absolute representations of these components. We evaluate our approach on four different musical understanding problems: monophonic voice separation, harmonic analysis, cadence detection, and composer identification which, in abstract terms, translate to different graph learning problems, namely, node classification, link prediction, and graph classification. Our experiments demonstrate that MusGConv improves the performance on three of the aforementioned tasks while being conceptually very simple and efficient. We interpret this as evidence that it is beneficial to include perception-informed processing of fundamental musical concepts when developing graph network applications on musical score data.

Sound,Artificial Intelligence,Machine Learning,Audio and Speech Processing

What problem does this paper attempt to address?

The problems that this paper attempts to solve are several key challenges in music understanding tasks. By proposing a new graph convolution block (MusGConv), which is specifically designed for efficiently processing music score data and inspired by general perception principles. Specifically, MusGConv focuses on two fundamental dimensions of music: pitch and rhythm, and considers the relative and absolute representations of these components. The paper evaluates the performance of its method on four different music understanding problems: monophonic part separation, harmony analysis, cadence detection, and composer identification. These problems can be transformed into different graph learning problems at an abstract level, such as node classification, link prediction, and graph classification. The experimental results show that MusGConv improves performance in the above three tasks while maintaining conceptual simplicity and efficiency. This indicates that it is beneficial to include perception - based processing of basic music concepts when developing graph network applications for music score data.