Sonic and acoustic links between real and virtual worlds in audio augmented reality

Bhattacharyya, Jacob D. (2026) Sonic and acoustic links between real and virtual worlds in audio augmented reality. PhD thesis, University of Glasgow.

Abstract

Audio augmented reality (AAR) applications enhance our real world surroundings with virtual audio. To truly augment our reality, AAR applications must be aware of those surroundings, however AAR applications currently have no understanding of a user’s aural environment. This thesis focuses on this fundamental gap in AAR, beginning by establishing a new definition for audio augmented reality, as a system that clearly connects the real world with virtual audio sources. It identifies “acoustic links” and “sonic links” as a way to create this connection and imbue AAR applications with an understanding of a user’s real world aural surroundings. Six mixed-methods user studies are presented, exploring the potential of these links.
This thesis first explores acoustic links, where the acoustics of a user’s surroundings inform an AAR system. Study 1 explores listener perceptions of different acoustic reproductions in a formal listening test. Study 2 builds on this to explore the plausibility of different acoustic reproductions in real-world spaces and when using AAR applications. The results from these studies suggest that even a simple reproduction of an environment’s acoustics is enough to create virtual audio that is plausibly realistic, facilitating an acoustic link between real and virtual.
Study 3 begins exploring sonic links, where the sounds in a real world environment inform an AAR system, by evaluating AAR applications which respond to human-produced sounds as control schemes. It compares existing AAR control schemes with novel sonic controls, finding that speech and “sonic gestures" are viable ways to control AAR applications and by extension create a sonic link. Study 4 explores the use of environmental sounds to create sonic links in three different AAR applications, finding that such environmental sonic links create a more augmented experience and heighten a user’s awareness of, and connection to, their real world surroundings. Study 5 builds on this by deploying existing sound classification models to drive these AAR applications, finding that existing detection models are capable of creating sonic links this way and maintaining this more augmented experience. Finally, Study 6 evaluates sonically linked AAR applications in uncontrolled, real-world scenarios over extended periods, finding that these sonic links continue to provide these benefits in the real world, can be facilitated now, and offer a stable experience over time.
The overall conclusion of this thesis is that acoustic and sonic links are viable ways to extend and enhance existing approaches to AAR applications, and are achievable using existing technology.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
• Colleges/Schools: College of Science and Engineering > School of Computing Science
• Supervisor's Name: Brewster, Professor Stephen and Vinciarelli, Professor Alessandro
• Date of Award: 2026
• Depositing User: Theses Team
• Unique ID: glathesis:2026-86082
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 26 Jun 2026 08:37
• Last Modified: 26 Jun 2026 08:39
• Thesis DOI: 10.5525/gla.thesis.86082
• URI: https://theses.gla.ac.uk/id/eprint/86082

Related URLs

• Enlighten Publications Record
• Article DOI
• Article DOI
• Data DOI
• Data DOI
• Data DOI