Contribution
Physics- and Psychoacoustics-Based Analysis of Viscoelastically Damped Membranes
* Presenting author
Abstract:
Viscoelastic Damping (VD), a type of internal damping characterized by its memory effect, i.e. the ability to store and reintroduce damped energy back into the system, plays an important role in materials like leather and polymers, often used in musical membranes, therefore influencing their sound. However, the relation between the parameters of VD and the perceptual characteristics of the resulting sound remains unclear. Unraveling this relation is, nonetheless, paramount: it could enable us to use VD as a sound-shaping element, fostering the development of new materials, interfaces and tools for musical expression. To this end, a physics-informed machine learning framework to analyze the physical relationship between the parameters of VD and the resulting frequency spectra was introduced in previous research. Building on those findings, this study presents a psychoacoustic analysis of a physics-informed dataset of a viscoelastically damped membrane, where a series of psychoacoustic features, identified in the literature as relevant to the perception of musical sounds, were statistically analyzed with regard to their variations in relation to systematic modifications in the parameter space of VD. Preliminary results are discussed, emphasizing the relationship between VD parameters and psychoacoustic features, as well as their implications concerning the perception of percussive sounds.