ABSTRACT. This article presents reflections on the artistic use of convolution in electroacoustic music composition, highlighting its potential to create hybrid, complex, and imaginary sonic spaces. The study examines works in which convolution with non-spatial or unconventional sources plays a central role in sound processing, with particular attention on how spatial perception emerges from timbral interaction and transformation. Three case studies are presented, referring to different compositional and performative practices: fixed media, score-based live electronics, and improvised electroacoustic performance. These applications are all conceived for a quadraphonic diffusion system. The paper is accompanied by a demonstration session including selected audio excerpts from the presented works.

ABSTRACT. Through its fusion of scientific inquiry and artistic experimentation, Carving Out Creativity invites people to experience creativity as a tangible link between minds, materials, and communities. The culmination of this process was spatial audio and immersive video installation of sonified and visualized electroencephalogram (EEG) data, which linked multiple locations across hundreds of kilometers through real-time, multi-sensory interaction via computer network. At each installation, visitors encountered a stone sculpture paired with its virtual counterpart in a distant location. Using a haptic interface, participants feel in real time contact made at another site - while being surrounded by visuals and sound created from stone carver EEG data during the carving process, offering a glimpse into how minds connect through artmaking. Dynamic visualizations of the brain data, along with 360 degree video documentation of the stone carvers and 3-D scans of carved stones, are layered to communicate the rich experience of making art in a shared creative setting. Immersive audio fills the space, consisting of EEG sonification mapped to an ambisonics-encoded 3-D accurate model of the EEG data collection cap and spatialized soundfield recordings. 32 channels of time series EEG data drove 32 individual synthesizers to represent the sensor activity. The piece comprises movements that correspond to the stone carving workshop process including six naturalistic experiences during which EEG measurements were taken: a pre-carving resting brain state, visioning, synchronized riffling, individual carving, critique, and a post-carving resting brain state. Some sections demonstrate the group average of EEG data, while other sections spatially represent each stone carver, up to four at a time, in their concurrent workspace.

ABSTRACT. This paper presents a rule-based pipeline that transforms a musical score into spatial automation for object- based rendering. A Compositional Rule Engine (CRE) reads a MusicXML file, extracts musical descriptors (dynamics and harmonic features), and maps them deterministically onto ADM- OSC spatial parameters (azimuth, elevation, distance). The rules operate across several levels, connecting and transforming the music’s formal structure into spatial behaviour. As a proof of concept, a case study on an original composition demonstrates the full workflow, from notation to object automation.

ABSTRACT. Natural soundscapes present unique challenges for spatial audio capture, rendering, and machine-learning applications due to the interplay of spatially extended stochastic textures and highly directional transients. Humans are sensitive to the spatial fidelity of these environments; yet, existing datasets typically offer lower-order Ambisonic or monophonic audio paired with video. This paper introduces the Waterside Audio-Visual Environments (WAVES-4K) dataset, a multimodal dataset comprising two hours and forty-two minutes of synchronised 6th-order Ambisonic audio and 4K video. Utilising a 64-channel spherical microphone array (Eigenmike EM64) and a Lumix GH5 or iPhone 15 Pro, we captured 17 audio-visual perspectives at 14 sites across three distinct UK topographies, each with natural water sources: the Lake District (characterised by rivers, lakes, and avian woodland calls), the Cornwall coastline (broadband sea, beach and river mouth textures), and the University of Surrey lakeside during the dawn transition. While the captures prioritised natural environments, the inclusion of intermittent human-origin sounds provides further data for research into robust sound event detection (SED) and source separation. The dataset is highly suited for benchmarking immersive rendering techniques, conducting psychoacoustic studies of nature-based presence, and developing multimodal machine learning models. We detail the field-recording method and provide an inventory of the captured scenes.

ABSTRACT. This paper presents a comparative analysis of ambisonic microphones of different orders for the measurement of spatial acoustic parameters in architectural spaces. Three ambisonic microphones - a first-order 4-channel Sennheiser Ambeo, a third-order 19-channel Zylia ZM-1, and a sixth-order 64-channel mh acoustics EigenMike em64 - were used to capture Spatial Room Impulse Responses in two concert halls and one church. The B-Format signals were decoded according to microphone-specific channel ordering and normalization conventions (FuMa/maxN for the Ambeo, FuMa/SN3D for the Zylia, and ACN/SN3D for the EigenMike), and the spatial parameters Early Lateral Energy Fraction (LF) and Late Lateral Sound Level (LJ) were extracted following ISO 3382-1. Results indicate systematic differences in values across microphones, attributable to differences in microphone order, decoder settings, and normalization methods. The influence of these factors on the reliability and consistency of spatial parameter measurements is discussed, with implications for standardization in architectural acoustics measurement practice.

ABSTRACT. Generated 3D scenes are increasingly available through photogrammetry, LiDAR, and neural scene representations, yet authoring the corresponding soundscapes remains an open problem. Although recent AI models can synthesize environmental sounds, spatial audio, and room responses, these capabilities are fragmented across different tasks and do not directly yield an immersive, navigation-aware soundscape for a 3D scene. We propose a layered soundscape-authoring framework for generated 3D scenes. The framework decomposes a soundscape into object/event audio, an ambient layer, and room impression, corresponding respectively to semantic congruency, environmental atmosphere, and acoustic plausibility. We organize existing AI and audio-processing methods according to these layers and integrate them into a soundscape design pipeline for generated 3D scenes. Through objective and exploratory subjective evaluations under controlled visual trajectories, we compare the layered design with scene-level AI generation baselines and layer-ablation variants. The results show that object/event audio is the main carrier of semantic and spatial congruency, suggest that directional ambience becomes important when foreground events are sparse, and indicate that our image-conditioned RIR instantiation exposes a gap between predicted room responses and the room impressions listeners expect from the visual scene. Based on these findings, we derive a gap taxonomy that identifies where current AI methods are usable, where human authoring remains necessary, and what capabilities are needed for future immersive soundscape generation.

ABSTRACT. Head-related transfer function (HRTF) is a key technology for achieving binaural spatial audio, and its personalized modeling is of great significance for enhancing the immersive auditory experience. However, traditional HRTF acquisition methods rely on complex measurement equipment, are costly, and the regression-based modeling methods have certain limitations in depicting the distribution of high-dimensional acoustic signals. To address these issues, this paper studies the generative HRTF modeling method. Firstly, a denoising diffusion probability model (DDPM) is constructed, using human measurement features and sound source direction as inputs, to generate the head-related impulse response (HRIR). To address the problem of many sampling steps and low efficiency in the inference stage of the diffusion model, this paper further introduces the Flow Matching method to accelerate the modeling of the generation process. By converting the traditional discrete multi-step denoising process into continuous-time probability path modeling, learning the velocity field from the noise distribution to the target distribution, the HRIR generation can be completed with fewer sampling steps, achieving a significant improvement in generation efficiency. Experiments are conducted on the public HRTF dataset, and the generated results are evaluated using metrics such as log-spectral distortion and inference time. The experimental results show that the HRTF reconstruction model based on DDPM has a much smaller parameter quantity and scale than the mainstream HRTF regression models; furthermore, Flow Matching can effectively reduce the sampling steps and significantly reduce the inference time (from 15 seconds to 4 seconds), improving the generation quality (from 5.13 dB to 4.80 dB), achieving a good balance between efficiency and performance.

ABSTRACT. The evaluation of multi-channel spatial audio systems and virtual auralization environments traditionally relies on subjective listening panels, a process that can introduce logistical bottlenecks and perceptual biases. This paper presents a preliminary, purely DSP-based evaluation framework designed to objectively score the psychoacoustic fidelity of high-density spatial rendering. To establish a mathematically stable baseline, the framework utilizes deterministic, algorithmically synthesized musical stimuli. By explicitly avoiding the use of external datasets, machine learning generation, or stimuli containing inherent noise or harsh spectral artifacts, the system isolates signal degradation introduced solely by the spatialization engine.

The evaluation methodology extracts key psychoacoustic features from the rendered spatial output. Specifically, the system calculates the resultant Roughness and Spectral Distance. To account for their distinct scales, these metrics are independently z-scored before being linearly combined into a unified objective perceptual score. While this metric integration is not intended to fully replace human perceptual evaluation, it provides a reproducible, scalable diagnostic tool. It allows researchers and audio engineers to computationally evaluate frequency coherence and phase stability across complex spatial trajectories, offering a rapid iteration cycle for optimizing spatial audio algorithms and acoustic digital twins.

ABSTRACT. In archaeological discourse, sound is frequently marginalized due to its ephemeral nature. However, for the ancient Greek world, the interconnection of sacred space, ritual, and sound was fundamental to the perception of the divine. This paper explores the transition from a purely visual investigation of Greek sanctuaries to a multisensory approach, focusing on the concept of "aural architecture" - the properties of a space that determine how it is experienced through listening.

Central to this research is the VOICING Project, which serves as a pilot initiative designed to bridge the gap between humanistic expertise (archaeology, history of religion, musicology) and technical acoustic research fields. By recognizing sacred buildings as "embodied spaces," the project investigates how architectural shapes, materials, and natural environments were intentionally or accidentally designed to enshrine sound, creating "earcons" - sonic events with profound symbolic meaning.

The methodology integrates cutting-edge digital tools to reconstruct these lost soundscapes. Through the analysis of two case studies the paper demonstrates the application of:

- 3D Modeling and Photogrammetry: To reconstruct the physical and environmental parameters of the sites.

- Auralization Techniques: Utilizing anechoic recordings of ancient instruments (such as the aulos) and choral performances.

- Virtual Reality (VR) Platforms: Developed in Unity for Oculus Rift, allowing researchers to experience the sites as "ancient listeners" and evaluate acoustic parameters (speech intelligibility vs. musical resonance) from various positions within the cultic "theatral steps."

By shifting the focus from quantitative acoustic data to qualitative "auditory space," this study contributes to the field of Digital Heritage and Soundscape Archaeology. It provides a framework for understanding how sonic events - from the rustling of leaves in a sacred grove to the rhythmic pulse of processional music - sacralized space and time, fostering social cohesion and immersive religious experiences in antiquity.

ABSTRACT. Abstract (short version) - This project proposes a unified framework for categorising, archiving, and labelling urban field recordings by associating each audio file with a structured metadata record. This metadata architecture integrates spatial, temporal, environmental, and technical descriptors compliant with internationally recognised standards. By using this system, we promote a collaborative model for building an extensive archive of acoustic documentation under a common protocol, with the aim to create a network of archaeological websites. As a case study, we conducted a series of multichannel and binaural recordings across historically significant sites of the city of Genoa, creating a digital repository that allows users to navigate the city through its acoustic atlas.

Abstract (extended version) - Urban field recordings have long served as an analytical tool for investigating a wide range of urban qualities, from environmental conditions to social dynamics. Decoupling sound from visual documentation provides insights which are relevant to several domains – including acoustic ecology, urban sociology, heritage studies and artistic practice – demonstrating how urban soundscapes are able to capture both the objective and the affective dimensions of public spaces. They also reveal the technological, social and cultural development of a city, along with its general identity. Given the significance of this shared interest, numerous attempts have been made over recent decades to collect audio recordings from a variety of contexts, many of which are urban, and to organise them into online archives that are often community-driven. However, while the quality of many of these initiatives is undeniable, there is currently no common method for the classification of urban audio documents, nor for sharing them across parties in such a way as to allow systematic comparison and analysis across different datasets. Most of the research fields for which urban acoustic documents are relevant would in fact benefit from the possibility of comparing them, both synchronically and diachronically. This can only be achieved if the recordings have been made, classified, and archived according to a common and methodical practice. This project proposes a unified framework for categorising, archiving, and labelling urban field recordings by associating each audio file with a structured metadata record. This metadata architecture integrates spatial, temporal, environmental, and technical descriptors compliant with internationally recognised standards, including Dublin Core (ISO 15836) for resource identification, ISO 19115 and GeoJSON for spatial data, ISO 8601 for temporal information, AES57-2011 for audio equipment metadata, and ISO 12913-1 for soundscape characterisation. It offers a model for systematic collection of urban audio recordings, while allowing for informed and specific searches in their archives. By using this system, we promote a collaborative model for building an extensive archive of acoustic documentation under a common protocol, enabling researchers, cultural institutions, and independent scholars to contribute and access data seamlessly. As a case study, we conducted a series of multichannel and binaural recordings across historically significant sites of the city of Genoa, both in the form of mobile soundwalks and fixed soundscapes. These audio recordings, together with the metadata files which encapsulate all the information concerning the recording location, time, context and techniques, have been consequently archived into a website, creating a digital repository that allows users to navigate the city through its acoustic atlas. This project was born and developed within the Electronic Music School of the Conservatoire Niccolò Paganini in Genoa. Drawing inspiration from the World Soundscape Project pioneered by R. Murray Schafer, it promotes a conscious listening practice and addresses the pressing need for dialogue and collaboration among institutions facing the rapid transformation of urban environments. Future generations of students will continue to add documented audio recordings to the website, ensuring the project's growth over time. This example, focusing on Genoa’s historical city centre, represents an attempt to satisfy both the systematic requirements of the academic researcher and the normal internet user’s desire for a light exploration of the acoustic characteristics and contexts of urban spaces. Ultimately, it serves as an invitation for other institutions to join this project, contributing to the collection of audio traces over time with the aim to create a network of archaeological websites where one may find the collective memory of urban lives.

ABSTRACT. An enhanced audio experience is becoming a re- quirement for many Augmented reality (AR) applications for greater immersion. To generate realistic 3D sound, a highly optimized HRTF model for mobile AR application is used. The model is efficient in terms of computational cost and uses GPS and compass sensors data to position the sounds as the user moves in real physical 3D space for real-time audio navigation. To evaluate the sound localization accuracy in an AR environment, this 3D audio engine is compared to the cosine panner model. The outcomes did not differ significantly between the two in regard to navigation. However, qualitative analysis indicates that this model should be used for AR applications. Through 3D sound, audio augmented reality enhances the reality perceived through the auditory system. This paper contributes to the advancement of this new technology. It will be used for creating various AR applications, including games and navigation, capitalizing on sound to enhance the immersive experience of 3D virtual environments and augmented reality.

ABSTRACT. Virtual acoustic systems are employed to simulate complex acoustic environments for auralisation. Generally such systems are disseminated over arrays of loudspeakers or binaural rendering. Acoustic sources and events are broadly modelled as infinitesimal points, positioned in the far field such that their impinging wavefronts are planar. The auditory cues conveying the shape and size of the source are omitted. Extending point-based rendering to include sources with non-negligible geometric size is a topic of great interest in the literature. Methods such as decorrelation and parametric distribution of multiple source copies are employed to spread the incident sound across a region in an attempt to convey its size, shape and distribution in space. Such systems are generally non-physical and use post-processing to achieve perceptually sufficient results. In this paper, a framework is presented drawing on the field of near-field spherical acoustic holography. The framework employs a physics-first approach, whereby we aim to render the binaural sound field produced by an arbitrarily shaped source with physical accuracy.

ABSTRACT. ATLAS (Actor Theater Localization Audio Space) is a real-time embedded spatialization system that continuously adapts the stereo panning and gain of a live performer’s amplified voice to their estimated stage position. Three Bluetooth Low Energy (BLE) beacons and a wearable ESP32 microcontroller implement real-time positioning; position data are forwarded over I2C to an ADAU1701 fixed-point DSP, which renders a position-coherent stereo output incorporating equalization, de-essing, and dynamic compression. The stereo signal is transmitted wirelessly to the theater’s existing audio mixer. Unlike professional object-based platforms, ATLAS is a wearable hardware, self-contained, battery-powered and operates without any sound-engineer intervention. The invention is protected under Italian Patent No. 102024000004768 (granted 09/03/2026), held by e-voluzione srl, Naples, Italy. More info on atlasaudio.eu.

ABSTRACT. The paper will describe the new Apple Spatial Audio Format (ASAF) and the new Apple Positional Audio Codec (APAC) that has recently been developed and deployed on Apple devices. ASAF comprises rich metadata, support for basic format atoms of channels, objects and Ambisonics as well as a corresponding adaptive renderer. The format was primarily developed to provide the realistic, compelling and immersive experience necessary to accompany visual experiences in Virtual Reality devices. Having met that bar, the format is also well positioned to provide more immersive experiences for traditional content such as music and TV episodic content. We will describe the philosophy, architecture and algorithms behind the new format as well the new codec that was specifically designed to stream these experiences.

ABSTRACT. This article details the collaborative process of developing and designing VOLTA, a portable loudspeaker array. The aim was to create a tool for smaller concerts, installations, and educational activities, primarily to disseminate works in ambisonics format and to provide the necessary facilities for the pedagogy of spatial audio in the Naples area, Southern Italy. The article focuses on the prototyping process, detailing the evaluations that led to the final loudspeaker design, the amplifier strategies, and an overview of costs. It also provides an overview of other relevant projects that inspired or informed the design of VOLTA, as well as information on subsequent plans and future design challenges.

ABSTRACT. The preservation and dissemination of cultural heritage are increasingly based on non-invasive digital technologies to avoid the risks of manipulation, environmental damage, and limited access to fragile objects. Three-dimensional (3D) digitization and additive manufacturing allow for the production of exact physical replicas, but most current applications only generate static reproductions with limited interpretive capabilities. This work experimentally proposes a non-invasive additive manufacturing framework for 3D replicas capable of interacting with them and providing synchronized audiovisual feedback. The application is carried out on the petroglyph resources of Miculla rock art in the Tacna Region, located in southern Peru. The study's workflow integrates the digital capture, physical fabrication, and sensory processes into a workflow that preserves the integrity of the original pieces. The results of the experimental evaluation demonstrate that the produced replicas maintain high morphological fidelity and allow for reliable interaction, low-latency audiovisual synchronization, and greater user participation. The results indicate that interactive replicas can serve as living interpretive interfaces, a scalable and conservation-focused solution for disseminating, educating, and making heritage accessible.

ABSTRACT. Hutchinson Synthesis is a parametric additive synthesis technique that organises the space of sound parameters as a structural analogue of Hutchinson’s N-dimensional hypervolume (1957). Time, frequency, amplitude, and spatial direction define the dimensions of the synthesiser’s Hutchinsonian niche. The system distributes sound events within niches defined by controllable parameter ranges for each dimension. Parameter distribution is governed by the GIGP (Generalized Immanantal Gibbs Process) framework, which unifies eight point process classes — pairwise interaction processes, stochastic intensity fields, temporal causal processes, and explicit geometric constraints — into three independent multiplicative blocks. The degree of separation and overlap between niches is continuously controllable per dimension through the parameter β. The spatial direction of sound sources is encoded via Higher Order Ambisonics up to seventh order, constituting in the proposed framework a dimension of the niche formally equivalent to the others. The system is designed as a generative tool for algorithmic composition, not as an autonomous compositional agent. Validation is conducted via Maximum Mean Discrepancy (MMD) across all eight implemented probabilistic distributions, including emergent intermediate regimes produced by combination of the GIGP blocks. Audio examples and reference configurations are available in the project repository.

ABSTRACT. This study evaluates the perceived binaural rendering quality of signals from various baffled microphone arrays, specifically spherical and equatorial microphone arrays with both spherical and non-spherical baffles, employing signal-independent rendering and magnitude equalization in the spherical harmonic domain. Using the multi-stimulus category rating paradigm, the arrays were compared to each other and several anchor conditions while omitting an explicit reference stimulus. Results reaffirm earlier findings, demonstrating substantial variance in subjects’ ratings of overall quality, with a retest variability of about 10% of the rating scale for repeated hidden anchors, which varied strongly between subjects with a supposed inverse relation to their listening experience. Among population-level effects that exclude individual response variance, the largest perceptual difference detected is about three times the variability, occurring between the low- and high-anchor conditions, with a high-accuracy spherical array achieving quality equivalent to a mannequin head. Direct evaluation of increasing the spherical harmonic order from 2 to 8 in equalized spherical and equatorial arrays confirmed only minimal improvements. However, comparison to an unequalized 4th-order array revealed substantial quality gains, highlighting the value of implementing a rigorously designed mid-anchor condition. For sources located in the equatorial plane, the different array types generally provide similar quality, although non-spherical arrays currently face drawbacks at lower array orders and during transient source excitation due to limitations in the rendering filters. Furthermore, the results indicate that drums exert minimal effects to prompt more critical evaluations of rendering conditions compared to source signals containing speech.

ABSTRACT. The intonarumori, experimental sound-producing devices developed by Luigi Russolo between 1913 and the early 1920s within the context of Italian Futurism, were conceived as instruments capable of reproducing and controlling the noises of the industrial and urban environment. The present study combines historical research, archival documentation to examine the sound production mechanisms and spectral characteristics of reconstructed intonarumori. Particular attention is devoted to the relationship between mechanical gesture and resulting sound morphology, highlighting the hybrid nature of these devices between musical instrument, machine, and sound effect generator

ABSTRACT. This paper presents the acoustic analysis of the Communal Theatre in Modena, Italy, and summarizes the realization of the traditional scenery by using tools like brushes and timber for the prompts. The measurements were analyzed to understand the acoustic response of the room in terms of T30, which was extracted across the octave bands between 125 Hz and 4 kHz. The paper discusses the birth of the elliptical shaped opera theatre as an alternative to the other plan layout that was developing between the 17th and 18th centuries, representing the effect of the Illuminist philosophy of finding rationalism behind arts. The results show that the reverberation time was found around 1.9 s, which is a good compromise between opera and symphonic music. Additionally, an overview of the traditional techniques regarding scenery has been included

ABSTRACT. This paper talks about the acoustic response measured within the Manifattura dei Marinati, in Comacchio, Italy. The acoustic survey was taken with different types of microphones, including a binaural dummy head, placed across the space according to the standard requirements of ISO 3382-1. The results show a good compromise in terms of definition between music and speech purposes. However, the reverberation time around 2.0 s may make speech understanding for listeners positioned in the last rows. Hence an amplified audio system would be beneficial to lower the arrival of late reflections and make the speech more intelligible.

ABSTRACT. The measurements of acoustic properties in Italian Stile Opera Houses are normally conducted with chairs in the stalls. In this paper, the Authors compare two theatres in which 3D acoustic measurements, conducted using a 32 channels mic array, were done without chairs in the stalls. The analysis compares 3D maps in similar positions and describes similarities and differences among the two theatres.

ABSTRACT. This study presents a comprehensive acoustic evaluation of the Church of Santi Gregorio e Siro in Bologna, a historical sacred space characterized by a complex evolution from the Renaissance to the Neoclassical period. Utilizing both traditional monaural/binaural parameters and advanced 3D spatiotemporal sound field mapping via a 64-channel spherical microphone array (Eigenmike), the research identifies a unique acoustic profile. Results reveal a long reverberation time (T20 of 5.3 s) and high early lateral energy levels (JLFc≈ 0.6). The church interior is characterized by long reverberation and relatively limited variation in reverberation characteristics across measurement positions. Speech intelligibility is critically low. These findings suggest favorable acoustic conditions for traditional sacred music and provide essential baseline data for future conservation and restoration.

ABSTRACT. This study presents a comprehensive acoustic investigation of the Oratory of San Filippo Neri in Bologna, focusing on the relationship between architectural configuration and sound-field characteristics in a reconstructed sacred space combining historical masonry and a contemporary wooden rib-vault structure. Particular attention is given to the contrast between the audience’s immersive experience and the performer’s perception at the elevated organ balconies. Measurements were conducted under unoccupied conditions in accordance with ISO 3382-1 and were complemented by 3D spatiotemporal sound-field mapping using a 64-channel spherical microphone array. The results reveal a mid-frequency reverberation time (T20) of 2.4 s, limited spatial variation in reverberation parameters, and high early lateral energy fractions (JLFc ≈ 0.5–0.8). The 3D maps further visualized the evolution of sound energy from early reflections to late lateral reverberation and revealed a concentration of reflected energy arriving from the direction of the stage area. By combining conventional room-acoustic parameters with advanced visualization techniques, this study provides a multifaceted understanding of the Oratory’s acoustic character and offers a possible physical basis for the performer’s perception of acoustic isolation at the organ balcony positions.

ABSTRACT. Expanding upon a 2024 preliminary survey, this study uses a multi-source measurement approach to characterize the acoustic field of the Church of SS. Salvatore in Bologna. Measurements followed ISO 3382-1 using monaural, binaural and B-format techniques, complemented by 3D spatiotemporal mapping with a 64-channel spherical microphone array. Mid-frequency EDT and T20 averaged 6.3 s and 6.6 s, respectively, with appreciable variation in EDT across measurement positions. Low clarity, definition and STI values indicate limited speech intelligibility, while high early lateral energy fractions indicate a substantial contribution of lateral reflections. The 3D maps provide complementary information on the arrival directions of reflected sound, including a concentration of late reflected energy arriving from the direction of the organ balcony under a specific source–receiver configuration. These findings document the church’s long reverberation, position-dependent acoustic characteristics and directional reflection patterns, providing a baseline for future conservation and further investigation of acoustic conditions for performers.

ABSTRACT. This study presents a comprehensive acoustic investigation of the Church of Santa Maria di Galliera in Bologna, conducted prior to a planned restoration to recover the historical color scheme on internal surfaces. Unlike the adjacent Oratory, this church preserves its Baroque architectural integrity and decorative elements. Acoustic measurements were performed in accordance with ISO 3382-1 alongside 3D spatiotemporal mapping via a 64-channel spherical microphone array. The results reveal a middle-frequency reverberation time of 3.9 s, which is relatively short compared to other historical churches of similar volume. The EDT/T20 ratio remained below 0.9, a decay profile consistent with those observed in some opera houses. The measured C80 values (−4 to −2 dB) suggest limitations in musical clarity, while D50 (20–30%) and the overall STI (0.4) indicate limitations in speech intelligibility. The 3D maps show changes in the arrival directions of reflected energy over time, complementing the spatial variations identified through conventional acoustic parameters. This study provides essential empirical documentation for future cultural heritage management and contributes to grounding subsequent restoration work in an informed understanding of the church's unique acoustic characteristics.

ABSTRACT. Immersive music has increasingly relied on advanced spatial audio technologies; however, its development remains largely centered on rendering techniques rather than on models that address the structural and perceptual role of space in musical processes. This paper proposes a framework for immersive music based on the notion of space as a structural parameter, understood not as a secondary attribute of sound but as an active component of musical organization.

The proposed model, termed Spatial Grammar, articulates spatial organization across four interrelated levels: parametric (continuous spatial fields such as position, distance, and distribution), functional (relational roles such as proximity, separation, and grouping), syntactic (temporal articulation of spatial configurations), and gestural (perceptually salient spatial events). Drawing on perceptual theories, including Gestalt principles and spatial hearing models, the framework emphasizes the listener’s experience as central to the structuring of immersive sound.

The paper also addresses a key practical challenge in immersive music: the translation of spatial structures across different audio systems, including Wave Field Synthesis (WFS), Ambisonics, and multichannel setups. Through the case study of Risonanze, a work originally composed in a WFS environment and later adapted to an 8-channel system, the paper examines how spatial relationships can be preserved or transformed when moving between fundamentally different rendering paradigms.

By integrating theoretical reflection and compositional practice, this contribution aims to shift the focus from spatial rendering to spatial thinking, proposing a model that supports both the analysis and creation of immersive music across heterogeneous technological contexts.

ABSTRACT. Hydroponic materials have recently been studied for use as food for future space missions and in arid regions of the planet where water is scarce or where soil is contaminated and adequate agricultural production is impossible. However, hydroponic materials were developed to produce acoustically effective materials. These materials are obtained by sowing wheat seeds on layers of various porous materials. The system is supplied with a small amount of water, allowing the seeds to germinate and grow. As they grow, the roots penetrate the underlying porous layer, creating an acoustically effective material. This paper reports measurements of hydroponic materials performed inside a reverberation chamber. The absorption coefficient values ​​obtained by diffuse incidence were inserted into a simulation model for architectural acoustics to evaluate their effective contribution to sound field control.

ABSTRACT. This work-in-progress study proposes an accessible pipeline for immersive audiovisual capture and reproduction in monumental architectural environments, framed as a multisensory Digital Twin. The approach integrates geometrical acoustics simulation and volumetric visual reconstruction to ensure perceptual coherence between sound and vision. Acoustically, a composite source model in Odeon 19 discretizes violin emission and encodes the resulting field in Higher Order Ambisonics (HOA). Visually, ongoing research explores emerging methods such as Instant4D and Diffuman4D for dynamic volumetric reconstruction and future 6DoF navigation. The system is currently implemented in Unreal Engine 5, with possible fallback to Unity for 4DGS integration. The study investigates how audio-visual synchronization influences perceptual realism and presence in immersive cultural heritage scenarios.

ABSTRACT. Auditory displays based on spatial audio technologies can complement or replace conventional graphical user interfaces when visual attention is limited. While such systems often treat sound images as interactive elements, the specific acoustic characteristics that make a sound image feel "suitable" for interaction remain unclear.

This study investigates the perceived suitability of virtual sound sources for button-press interaction, to explore effective acoustic parameter combinations for rendering “acoustic buttons.” Our system renders a focused sound source in front of the user using wave field synthesis with a 64‑channel linear loudspeaker array. The focused sound source and its surrounding interaction region together constitute an "acoustic button." Fingertip positions captured via motion tracking are used to determine interaction states such as contact and press.

We conducted a subjective experiment to examine the factors that influence the perceived suitability of such interactive sound images for button pressing. We investigated multiple factors related to acoustic button design, including (1) action-contingent auditory changes (e.g., amplitude and spatial attributes), (2) autonomous movement along the depth axis, and (3) a variety of sound stimuli. A series of experimental conditions were designed to evaluate how these spatial and behavioral factors influence the perceived suitability of the "acoustic button."

The results indicate that significant differences were observed in how naturally a sound image is perceived as suitable for button-press interaction across sound-source types and autonomous movement conditions. Responsiveness to user actions and perceived ease of pressing appeared to be related to temporal structures within the stimuli, particularly the duration of silent intervals. Additionally, the perceived suitability for pressing may be reinforced by prior associations with familiar operational sounds, such as the common experience of interacting with alarms. These findings offer practical acoustic design principles for rendering intuitive and effective interactive sound images in 3D auditory environments.

ABSTRACT. For many centuries, circular and vaulted structures were realized and used for aesthetic and constructive needs as buildings of worship, monumental buildings or courtyards. In vaulted environments, the effects of sound reflection focus on the central area due to the focus of multiple reflections. This paper presents the results of acoustic measurements carried out in a circular plant environment with a vaulted ceiling. In the central area of the rooms, due to the geometry of the environment, there are focusing sound effects that generate a “flutter echo”.