ABSTRACT. Previous research in virtual reality has shown that changing the visual appearance of a user’s virtual avatar can alter the user’s own body representation. In contrast, comparatively little is known about whether such transformations can be induced solely through auditory feedback that is spatiotemporally synchronized with voluntary body movements. This study investigates whether a sense of having wings on one’s back—or of one’s arms being wings—can be elicited by rendering spatially dynamic wing sounds that follow the user’s flapping motions. We developed an interactive audio system that employs a 42-channel spherical loudspeaker array to render real-time moving virtual sound sources synchronized with the user’s arm movements tracked via motion capture. Spatial rendering was implemented using Near-Field Compensated Higher-Order Ambisonics (NFC-HOA) with IIR filters. When the user performs horizontal arm opening and closing motions, virtual sound sources are positioned either behind the user or at the arm locations. The wing sounds are synthesized by modulating the gains and spectral characteristics according to arm velocity and acceleration. A subjective evaluation was conducted to assess whether the proposed system could induce a sense of wing ownership. To examine the influence of sound source distance and direction, we compared NFC-HOA rendering with two conditions: Higher-Order Ambisonics (HOA) rendering without distance control, and stereo reproduction using loudspeakers placed either behind or in front of the listener. The results showed that a sense of having wings on the back was rarely induced under any condition when sound sources were rendered behind the listener. In contrast, when sound sources were rendered at the listener’s arm positions, the results suggest that a sense of the arms being wings was induced under both the NFC-HOA and HOA conditions. These findings suggest that presenting spatially localized auditory feedback at the actual position of a moving body part can contribute to inducing altered body-part representation, even in the absence of visual feedback.

ABSTRACT. This paper presents a hybrid spatial audio approach based on a case study integrating airborne multichannel diffusion and bone-conducted audio within a unified perceptual system. While spatial audio research has traditionally focused on loudspeaker-based rendering techniques, the coexistence of multiple propagation modalities remains underexplored. The proposed work investigates the interaction between an external spatial sound field and an internal, body-mediated auditory layer delivered through bone conduction. The system is implemented as an adaptive installation in which listener movement dynamically influences the relationship between the two layers. Particular attention is given to calibration strategies aimed at reducing perceptual masking through spectral and temporal differentiation. Observations from a public presentation suggest that the combination of short, high-frequency bone-conducted events and diffuse multichannel backgrounds supports perceptual separation while enabling a continuous transition between individual and collective listening. From this implementation, a set of design strategies for managing coexisting propagation modes is derived.

ABSTRACT. CConventional speech-in-noise clinical tests often fail to reflect the multisensory and dynamic nature of real-world speech reception in noise. This study presents an immersive platform aimed at getting speech reception evaluation closer to real-world conditions. To this scope an audio-video test bench was developed that enables an accurate spatial rendering of complex acoustic scenes with realistic visual ambience. A custom application synchronizes auditory stimulus delivery through multichannel audio, immersive visual scenes presented through a virtual reality headset, and the real-time collection of behavioral and physiological data. Speech reception accuracy is obtained by machine learning processing of the listener’s vocal output and the system acquires response time, pupillometric data, skin conductance and heart rate. The platform has been conceived for applications in a clinical-audiological context; five realistic audio-visual scenarios have been developed which are characterized by different listening demands and background noise conditions, both indoors and outdoors. The scenarios are created by merging synchronized 360° video and multichannel audio recordings. Speech reception accuracy is assessed using the Italian Matrix Sentence test, presented in the open-set format. This immersive platform is expected to provide a more comprehensive and ecologically relevant characterization of speech intelligibility in noise, supporting the adoption of multimodal assessment protocols for both research and clinical applications.

ABSTRACT. The interpretation of fixed-media electroacoustic music involves processes of historical revision, technical updating, and perceptual reformulation. Based on the spatial reinterpretation of Luciano Berio’s Thema (Omaggio a Joyce) and Visage, this paper examines the potential of immersive audio for reworking stereophonic works. The aim is not simply to enlarge the listening field, but to render explicit spatial tensions and directionalities arguably embedded in the sonic material, yet constrained by the original stereo format. The methodology combined comparative listening of restored versions and previously realized multichannel interpretations with detailed analysis of the source material and the design of spatialization strategies based on differentiated-order Ambisonics encoding, combining low- and high-order representations. A key stage involved the use of source-separation and spectral-unmixing tools to isolate signal components and assign them to distinct spatial positions or trajectories. The case study suggests that enveloping audio can operate not only as a means of spectacularizing space, but as a critical tool for reactivating the complexity of a work. Spatialization thus emerges as a form of critical reading of the electroacoustic repertoire, where restoration, spectral separation, and spatial distribution converge to produce a new perceptual ecology of listening.

ABSTRACT. Heritage sites frequently exist as a collection of fragmentary material and documentary evidence, limiting the possibility of reconstructing accurately and comprehensively their original acoustic conditions. This challenge is evident at the sacred center of Coricancha (Cuzco, Peru), where extensive colonial-period and modern transformations obscure the spatial configuration and architectural makeup of the complex as it stood when it was built during the Inca Empire (\textit{ca}. 1440–1532 C.E.). While these alterations pose an obstacle to acoustic research, they do not prevent it altogether. This paper addresses how, under conditions of restricted access and partial data, acoustic measurements can support bounded interpretations of past sonic environments and how these interpretations can be meaningfully communicated. The approach adopted is grounded in \textit{in situ} impulse response measurements collected across accessible areas of the site. Acoustic parameters, including sound pressure level, clarity coefficient, and frequency response, are analyzed to characterize patterns of audibility, intelligibility, and spatial differentiation. Rather than attempting a speculative sonic reconstruction of Coricancha, these measurements are treated as empirical constraints that reveal zones of acoustic continuity and separation, offering insight into how the imperial Inca built environment as it was developed at this sacred center may have structured the contemporary sonic experience. To extend these findings beyond static analysis, we develop a minimal web-based interface that enables interactive exploration of the measured acoustic field that can be used by specialists and the public at large.

ABSTRACT. This paper presents a tangible hemispherical controller that captures spatial orientation as a single, intuitive gesture and integrates it into established ambisonic mixing workflows. It describes the hardware design and implementation, software mapping, and its integration into conventional mixing tasks. Comparing against currently available interfaces for spatial audio production, this work demonstrates an interface that simplifies the spatial arrangement of sound sources. We discuss a customizable integration into existing digital audio workstations using open source plugins and the open sound control protocol, a strategy for spatializing reverb in relation to its dry source, and a set of automation writing features using integrated gestures.

ABSTRACT. This paper presents an interdisciplinary methodological protocol for the digital reconstruction of ancient theatres, developed as part of the Voicing Spaces project, with a particular focus on integrating geometric accuracy, architectural interpretation and future acoustic simulation. The research focuses on a case study of the Morgantina Theatre (central Sicily), combining aerial and ground-based photogrammetry, and georeferenced surveying techniques to produce consistent and metrologically reliable datasets. These data are interpreted through a critical architectural analysis aimed at identifying the main structural components and supporting scientifically grounded reconstruction hypotheses. The resulting 3D models are semantically organised and implemented within VR tour platforms and real-time XR environments developed in Unreal Engine, designed as flexible and updatable systems. The proposed methodology ensures continuity throughout all stages of the process and establishes a scalable infrastructure for the future integration of spatial audio simulation, enabling immersive and acoustically informed explorations of ancient theatrical spaces.

ABSTRACT. This abstract presents Resonances of Stones, an immersive architectural and acoustic reconstruction of the Bazeos Tower in Naxos developed as a case study in acoustic heritage. The project combines in situ First-Order Ambisonics impulse-response capture, acoustic analysis, binaural auralization, and interactive audio-visual XR deployment in Unity and Wwise. RTPC-driven interpolation between zone-based impulse responses, (convolution reverb) and dynamic early reflections support a 6DoF listening experience. The heritage dimension is grounded in a musicological framework that interprets the tower as a stratified cultural space and links its historical, spiritual, and social identities to the selection and reinterpretation of Greco-Byzantine poetic-musical sources. The paper proposes a method connecting immersive audio technologies, musicological interpretation, and participatory heritage reactivation for contemporary and future audiences.

ABSTRACT. Archaeoacoutic research is producing a rapidly growing collection of spatially organized impulse responses. However, making these data sets accessible beyond a small group of specialists remains a design-challenge. This paper examines how interface choices shape the ways acoustic heritage can be explored, compared, and interpreted. Drawing on four archaeological and heritage case studies---Chauvet-Pont-d'Arc Cave in France, sites in the Bears Ears region of Utah, and Coricancha and Carpa Uasi in Peru---we describe three web-based interface models developed for different dataset structures and research questions. CAVERN supports discrete comparison between documented acoustic positions; ARIADNE enables movement along researcher-defined paths through spatially distributed measurements; and LiLab uses guided listening activities to foreground specific acoustic contrasts. Across these systems, we identify recurring design priorities for interactive archaeoacoustic access: formalizing impulse responses as spatial data, selecting navigation modes according to the structure and evidential limits of the dataset, reducing technical and geographic barriers to access, directing attention toward acoustically relevant distinctions, and maintaining transparency about what has been measured, interpolated, or reconstructed. We argue that greater data density, interaction freedom, or visual immersion do not necessarily produce more informative listening experiences. Instead, interface design forms part of the methodological translation between acoustic fieldwork, archaeological questions, and public engagement.

ABSTRACT. Binaural beats (BBs) arise from central auditory processing when two sinusoidal tones with slightly different frequencies are presented dichotically, resulting in the perception of a modulation corresponding to the interaural frequency difference (delta f). Although BBs have been investigated extensively in psychoacoustic and applied research, reported effects remain inconsistent, possibly because the perceptual limits and reliability of BB perception are still insufficiently characterized. The present pilot study aimed to replicate and extend the classical work of Licklider et al. (1950) using contemporary psychoacoustic methods. Seven normal-hearing participants completed two BB detection tasks—a method-of-adjustment task and an adaptive staircase procedure—in a within-subject test–retest design. Carrier frequencies ranging from 125 to 1412 Hz were investigated, and the maximum detectable delta f was determined for each carrier frequency. In addition, participants provided subjective comfort ratings for the individually adjusted BB stimuli at the just perceivable delta f. Results revealed substantial interindividual variability and moderate test–retest consistency across both tasks. While carrier frequency influenced BB perception, the classical detection pattern reported by Licklider et al. was not consistently reproduced across participants. Subjective comfort ratings decreased with increasing carrier frequency, with higher-frequency stimuli generally rated as less comfortable. These findings suggest that BB perception is less robust than often assumed and may depend strongly on listener-specific factors and prior experience. Improved psychoacoustic standardization and assessment of perceptual reliability may therefore be essential for future studies investigating cognitive or affective effects of binaural-beat stimulation.

ABSTRACT. Proximity voice chat is widely used in online games but remains under-explored academically. Existing research has focused on social interaction and moderation, with limited literature observing how modern advancements in audio simulation have impacted the feature. This research aims to explore the impacts of ray-traced environmental acoustic simulation applied to multiplayer voice communication. Within-subject testing utilized a prototype game with participants (n=14) in a controlled LAN environment. 12 of the 14 participants reported a preference for the environmental effects and observed behavioral responses suggest that enriched spatial information altered player interaction, including emergent ‘Marco, Polo’-style vocal search behavior. Participants reported enhanced perceived vocal realism, increased immersion and engagement, and improved spatial awareness. A measurable performance cost was observed, particularly in frame-time consistency, with substantially worse low-percentile frame rates when simulating effects. However, within the context of this simplified prototype, the performance stayed within acceptable bounds. These findings suggest that environmental audio effects in proximity chat improve perceptual presence and environmental interpretation in multiplayer play. These results support the conceptualization of vocal audio as gameplay audio, rather than solely a communicative tool.

ABSTRACT. Subjective assessment of room acoustic quality requires reliable listening conditions and clearly understood perceptual descriptors. this paper presents a revised questionnaire for evaluating acoustic quality in virtual environments, developed from earlier surveys of italian opera houses and selected items from the ircam questionnaire. the methodology addresses differences in listeners’ musical experience and the semantic variability of acoustic terminology across languages. a preliminary questionnaire assesses respondents’ understanding of relevant descriptors, while the main questionnaire incorporates ratings of item importance and records test completion time as an indicator of response reliability. a graphical user interface enables real-time switching between reproduction configurations and room impulse responses, allowing comparative assessments using anechoically recorded music. the experimental framework draws on the arlecchino and agorà listening rooms at the university of bologna, incorporating ambisonics, stereo dipole and wave field synthesis technologies. a database of binaural and b-format impulse responses from historic italian opera houses and international concert halls supports the virtual reconstruction of acoustic environments. the proposed approach provides a methodological basis for investigating relationships between subjective judgements, measured acoustic parameters and sound reproduction methods, while accounting for listener characteristics and linguistic differences.

ABSTRACT. The progressive electrification of residential heating and cooling systems is expected to significantly increase the diffusion of air-source heat pumps within urban environments. Although heat pumps contribute to energy decarbonization objectives, the widespread installation of outdoor units may produce non-negligible impacts on the urban acoustic environment. This study presents a numerical investigation aimed at quantifying the effects of extensive heat pump deployment on urban soundscapes under different urban morphologies and installation configurations. Advanced outdoor sound propagation simulations were performed using the Wölfel IMMI 2023 software platform. Four representative Italian urban districts were selected as case studies, characterized by different building densities, spatial distributions, and building heights. The analysed urban grids covered areas of approximately 75,000 m² and included residential contexts ranging from low-density suburban districts to compact high-density urban fabrics. For each building, geometric and volumetric properties, number of housing units, and estimated thermal demand were determined through GIS-based reconstruction methods using GEOscopio and Google Earth data. Heat pump thermal capacities were assigned proportionally to residential volumes, while sound power spectra were derived from experimentally characterized units grouped by thermal capacity ranges. A total of 24 simulation scenarios were developed by combining four urban layouts with six installation configurations. The analysed configurations included both decentralized systems, with outdoor units associated with individual apartments, and centralized systems with rooftop or ground-level installations. Simulations assumed continuous full-load operation, omnidirectional point-source radiation, reflective building façades, and receiver heights representative of pedestrian exposure conditions. The results demonstrate that both urban morphology and installation strategy strongly influence the resulting acoustic impact. Dense urban districts characterized by narrow street canyons and high building concentrations exhibited the highest spatial extent of elevated sound pressure levels. Decentralized configurations with multiple façade-mounted units generated widespread medium-to-high noise levels, whereas centralized rooftop solutions generally reduced pedestrian-level exposure despite localized increases near buildings. In several scenarios, areas exposed to sound pressure levels above 50 dB(A) occupied a significant fraction of the modelled urban grid, particularly in compact urban fabrics and centralized ground-level configurations. The study highlights the importance of integrating acoustic criteria into future urban energy transition strategies. The proposed methodology provides a useful framework for evaluating the acoustic sustainability of large-scale heat pump deployment and supports the development of urban planning, architectural, and plant-design guidelines aimed at mitigating adverse effects on urban soundscapes.

ABSTRACT. This study investigates the acoustical behavior of the Teatro Olimpico in Vicenza, the oldest permanent indoor theatre of the modern era. Designed by Andrea Palladio in 1580, the venue is uniquely characterized by its fixed wooden perspective scenery, installed by Vincenzo Scamozzi in 1585. The experimental campaign involved the acquisition of Impulse Responses (IR) across multiple locations within the cavea. An omnidirectional sound source was strategically positioned both on the proscenium and within the wooden stage scenery, with signals captured via measurement microphones to simulate the auditory experience of spectators in different seating areas. Furthermore, binaural analyses were conducted using a Head and Torso Simulator to provide a high-fidelity characterization of human spatial perception. Recognizing that the appreciation of a drama or concert is a multisensory experience influenced by acoustic stimuli, visual cues, cultural background, and emotional factors, the future developments of this research aim to quantify the impact of "non-acoustical" parameters on perceived sound quality. This holistic approach seeks to better understand the complex interplay between the physical environment and the subjective listener experience in heritage theatres.

ABSTRACT. Using simulated, artifact-free spatial filtering in an interactive, immersive virtual acoustic reality enables the spatial and spectral effects of signal enhancement to be disentangled when analyzing human behavior. This study validates a fully simulated beamformer within the interactive virtual acoustic rendering system TASCAR. We compared its performance with that of a minimum variance distortionless response (MVDR) beamformer with sampled microphone signals, and an MVDR beamformer with simulated microphone signals, both implemented in the openMHA framework. The fully simulated beamformer dynamically steers its direction in real time based on the participant’s head orientation. It applies broadband attenuation depending on the angular distance between the head direction and the direction from which sound sources arrive. Image sources are treated the same as direct sources with the correct image source position to ensure consistent spatial filtering across the acoustic scene. In the rendering system, diffuse sound fields are simulated using first-order Ambisonics. This requires a dedicated spatial filtering approach based on modal beamforming in the proposed filter. Performance was assessed via beam pattern analysis and diffuse sound field attenuation across identical acoustic scenarios. The results revealed systematic differences in directional selectivity and diffuse sound field attenuation between the fully simulated and the MVDR beamformers, which can be attributed to the frequency dependency of the MVDR beamformer.

ABSTRACT. Acoustic simulation provides a promising avenue to facilitate a software approach that closely resembles the real-life task of arranging instruments and microphones in a room. However, simulations typically take a long time, especially on consumer-grade hardware. The objective of this work is to use deep learning to reduce the geometric complexity of rooms, represented as point clouds, to alleviate the computational load of acoustic simulation without sacrificing the perceptual accuracy of the output. We examine an existing approach of reconstructing point clouds with an autoencoder and adjust the model architecture to improve results for indoor spaces. We devise an autoencoder capable of smoothing room geometries and provide the novel loss function required to do this; crucially, our autoencoder incorporates a novel acoustic loss term which ensures that perceptually important features are undisturbed, whilst perceptually irrelevant features are smoothed. We investigate the relationship between smoothing and acoustic integrity by adjusting importance hyperparameters. We find that our model smoothes perceptually irrelevant features and retains acoustic integrity, producing an average sound ray direction discrepancy that reaches $6.470^\circ$; this value indicates perceptual success. Further, we investigate the wider applicability of our model by applying our model to the task of acoustic panel placement and assessing our method in the context of digitised historical spaces.

ABSTRACT. This paper introduces an affective audio rendering framework that dynamically links human microexpressions to real-time 3D soundscape generation in immersive environments. While prior work in affective computing has primarily focused on visual adaptation or explicit user input, the integration of subtle, involuntary facial signals—microexpressions—into auditory rendering remains underexplored. We propose a closed-loop multimodal system that captures high-frequency facial data, extracts microexpression features using computer vision models (e.g., OpenFace), and maps these signals to spatial audio parameters such as source localization, reverberation, spectral balance, and temporal dynamics via the Web Audio API and spatialization engines.

The core contribution lies in defining a bidirectional coupling between affective state and auditory perception, where microexpressions continuously modulate the acoustic environment, and, in turn, the evolving soundscape influences user emotional responses. We formalize this interaction through an affect-to-audio mapping model grounded in valence–arousal space and implement it within a WebXR-compatible prototype supporting head-tracked spatial audio. A controlled user study evaluates the system across varying levels of audio immersion and responsiveness, measuring microexpression frequency, latency, perceived presence, and cognitive load.

ABSTRACT. In 1849, the first of three theaters designed by the papal architect Luigi Poletti, a native of Modena, opened in the city of Terni, Italy. A few years later, the Teatro Nuovo in Rimini (1857), subsequently named after Amintore Galli, and the Teatro della Fortuna in Fano (1863) were inaugurated. The theater in Terni was severely damaged during the Second World War and subsequently demolished. Following a lengthy debate and an international competition, construction of the new Teatro Verdi began in 2022. The new building is entirely different from Poletti’s original theater. This paper describes the main objectives guiding the acoustic quality design of the new theater.

ABSTRACT. The demand for immersive listening experiences has significantly increased in recent decades, largely due to the rapid development of the gaming industry, which has driven research toward the creation of stable and realistic Higher-Order Ambisonics (HOA) systems combined with highly immersive visual environments. This growing interest has encouraged research institutions and audio laboratories to equip themselves with dedicated listening rooms for experiments in spatial audio reproduction and sound localization. This paper presents the design of a demountable listening room, conceived as one of the main objectives of the AGORA project. The proposed system consists of 25 Ambisonics loudspeakers mounted on a temporary steel structure designed to reproduce immersive audio through Higher-Order Ambisonics (HOA) techniques, in addition to a ring of Wave Field Synthesis (WFS). The listening room is installed in Tresigallo, Italy, and already open to the public, offering visitors the opportunity to experience urban soundscapes recorded in different environments.

ABSTRACT. The study develops an acoustic concept for converting the former Sacro Cuore church into an auditorium, principally for small orchestral ensembles, soloists and choirs, while retaining the possibility of conferences and spoken events. Its preferred solution, C02, combines reflecting and diffusing panels, a suspended diffuser above the nave and selective acoustic plaster. The objective is a more reverberant, musically supportive room with a balanced sound field and fewer local focusing effects.

ABSTRACT. Collaborative work across geographically distributed studio environments presents a persistent challenge: how can artists, researchers, and technicians develop a meaningful sense of a remote space's acoustic character before physically visiting it? Existing approaches range from high-fidelity digital twins — requiring extensive geometric reconstruction, material modeling, and acoustic simulation — to simple documentation that fails to convey spatial experience. Neither serves the practical needs of a consortium of diverse labs seeking to share, compare, and creatively engage with one another's spaces.

This Demo presents a browser-based platform for constructing navigable virtual acoustic environments, developed by Fränk Zimmer and Sounding Future in collaboration with the Spatial Sound Consortium, founded and directed by Victoria Keddie. Built on Three.js for three-dimensional spatial rendering and ATMOKY's engine for binaural and object-based audio processing, the framework proposes a middle path: the virtual build. This spatial preview environment conveys meaningful acoustic impression and visual orientation without the overhead of full simulation. Browser-based delivery is a foundational design decision that removes installation barriers with historically limited access to spatial audio tools. This allows for a practitioner at an independent studio in Chicago and a researcher at a university lab in Belfast to inhabit the same virtual acoustic space without bespoke hardware or software licensing.

The acoustic dimension is grounded in a standardized impulse response protocol. Participating labs provide measured IRs at predefined listening positions, enabling convolution reverb to apply each space's acoustic fingerprint within the virtual environment. The visual dimension supports multiple levels of fidelity — from simplified .glb geometry or 360° photography to Gaussian splatting — prioritizing spatial intuition and recognizability over photorealism. Users navigate in both first- and third-person views, balancing immersive listening with system-level overview.

The system supports multiple simultaneous users within a shared three-dimensional environment. Each participant can navigate spatial configurations and perform within a common acoustic architecture, which is experienced differently from each listening position. This asymmetry within a shared space is a compositional condition the tool is designed to make productive, reflecting the Consortium's understanding of spatial audio as inherently relational.

Development is co-driven by core members of the Consortium's studios, spanning large-scale multichannel speaker arrays to compact binaural listening environments. Member studios serve as both pilot sites and active co-creators: they do not adapt their practice to an existing tool but directly shape its architecture through iterative feedback. A standardized test piece anchors structured listening sessions across sites, providing consistent benchmarks across diverse configurations. Feedback informs interface design, audio engine parameters, and spatial rendering logic. Key technical challenges addressed include latency management across distributed connections, consistent binaural rendering across varied headphone hardware, and building an interface that supports compositional thinking without requiring specialized technical training.

Early development outcomes include multi-user sessions and position-based IR interpolation, early indicators of what distributed spatial performance can look like when built on emerging browser-based technologies. The platform sits within a broader connected ecosystem: a global studio mapping component, nine physical member sites, and a travelling live performance program. No existing platform was designed to hold all of this. These pilot findings are the first evidence of what is possible.