Modelling the aesthetic response to audio signal processing, with reference to spatial sound reproduction
Loading...
Date
2026-06-18
Authors
Μοιράγιας, Γιώργος
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This dissertation investigates the factors that shape listeners’ aesthetic response in spatial audio reproduction and proposes methodological and modeling frameworks that integrate cognitive, perceptual, affective, and temporal aspects of listener experience. Conventional audio evaluation methods typically rely on static perceptual ratings and emphasize isolated attributes of sound reproduction. However, recent advances in immersive and interactive audio technologies highlight the need for more comprehensive approaches capable of capturing the multidimensional and dynamic nature of listener experience. In spatial audio environments, this experience is influenced not only by the acoustic properties of the reproduced sound field but also by listener expectations, emotional engagement, and the temporal evolution of signal attributes. To address these challenges, this work develops experimental methodologies and signal/channel-based models that enable a deeper understanding and improved prediction of listener responses in spatial audio systems. The dissertation is structured as follows:
Chapter 1 introduces the research topic and discusses the motivation behind the present work, situating the study within the broader context of spatial audio perception and listener experience research. The chapter defines the main research objectives, summarizes the key contributions of the dissertation, and provides an outline of the thesis structure.
Chapter 2 presents the theoretical background and methodological framework adopted throughout the dissertation. It reviews key concepts related to spatial audio perception and listener experience, addressing both perceptual and affective attributes, and introduces the evaluation methodologies employed in the listening experiments. These theoretical and methodological foundations provide the basis for the experimental procedures and modeling approaches developed in the subsequent chapters.
Chapter 3 introduces an initial analytical framework for modeling Overall Listening Experience (OLE), investigating the relationship between cognitive listener preference and perceptual attributes related to the reproduction of binaural audio signals.
Chapter 4 extends the previously introduced framework by incorporating affective listener responses, investigating how emotional dimensions, such as valence and arousal, contribute to the formation of listener preference in spatial audio reproduction. By integrating emotional responses with cognitive and perceptual attributes, this chapter provides a more comprehensive perspective on the factors that shape Overall Listening Experience.
Chapter 5 proposes a novel methodology for the temporal assessment of perceptual sound attributes, enabling the dynamic evaluation of spatial perception during listening. This methodology provides a framework for analyzing time-varying aspects of auditory perception that are not adequately represented by conventional static evaluation methods.
Chapter 6 develops machine learning models for predicting the temporal evolution of perceptual attributes. Building upon the previously introduced temporal assessment methodology, the proposed models examine the relationship between acoustic signal and channel features and the resulting time-varying perceptual responses of listeners.
Chapter 7 summarizes the main findings of the dissertation and discusses directions for future research in the modeling and evaluation of listener experience in spatial audio systems.
Description
Keywords
Perceptual audio, Acoustics, Aesthetic, Spatial sound, Temporal evaluation