Musical analysis of acousmatic music with psychoacoustic descriptors

This article presents an exploratory study in musical analysis with a focus on perceptual features of the sound. The development of computer science of the last century afforded the study of perceptual models through technological devices, which eventually became supports for music creation and analysis. Current studies focus on the application of computational tools in the context of systematic musicology. To study two acousmatic pieces composed by the authors, the text updates this discussion based on a psychoacoustics theoretical framework. We present the compositional sound material, focused on speculations about equal-tempered scales and its psychoacoustic consequences based on roughness models. To focus on aspects of musical texture, we use the following audio descriptors: the spectral flux, as a tool to represent microtemporal phenomena, such as roughness; the bark coefficients, to understand the development of texture in the frequency domain described by a perceptive model; and sharpness, to represent a quantitative measure of perception of density. As a result, we present graphical of the descriptors, with contextualized and comparative analyzes of the two pieces. We analyze and interpret the sound features described in this study as an emergent property of the sound texture. Finally, we discuss the potential of new perspectives for musical analysis based on perceptual models.