AES Show 2024 NY

Road tunnels are subject to strong regulations and high life-safety standards. Requirements importantly include the speech intelligibility of the emergency sound system. However, a road tunnel is an acoustically challenging environment due to extreme reverberation, high noise floors, and its tube-like geometry. Designing an adequate sound system is a challenge and requires extensive acoustic simulation. This article summarizes recent design work on several major road tunnel projects and gives practical guidelines for the successful completion of similar projects. The project includes several tunnels, each of several kilometers’ length with one to five lanes, transitions and sections, having a total length of 33 km. For each tunnel, first a working acoustic model had to be developed before the sound system itself could be designed and optimized. On-site measurements were conducted to establish data for background noise including jet fans and various traffic situations. Critical environmental parameters were measured and reverberation times were recorded using large balloon bursts. Sprayed concrete, road surface, as well as other finishes were modeled or estimated based on publicly available data for absorption and scattering characteristics. To establish the geometrical model, each tunnel was subdivided into manageable segments of roughly 1-2 km in length based on theoretical considerations. After calibrating the model, the sound system was designed as a large number of loudspeaker lines evenly distributed along the tunnel. Level and delay alignments as well as filter adjustments were applied to achieve the required average STI of 0.48. Validation by measurement showed good correlation with modeling results.

Noise emission into the neighborhood is often a major concern when designing the configuration of an open-air sound system. In order for events to be approved, advance studies have to show that expected immission levels comply with given regulations and requirements of local authorities. For this purpose, certified engineering offices use dedicated software tools for modeling environmental noise propagation. However, predicting the radiation of modern sound systems is different from classical noise sources, such as trains or industrial plants. Sound systems and their directional patterns are modeled in electro-acoustic simulation software that can be fairly precise but that typically does not address environmental issues.
This paper proposes to use a simple data exchange format that can act as an open interface between sound system modeling tools and noise immission software. It is shown that most immission studies are conducted at points in the far field of the sound system. Far-field directivity data for the sound system is therefore a suitable solution if it is accompanied by a corresponding absolute level reference. The proposed approach has not only the advantage of being accurate for the given application but also involves low computational costs and is fully compliant with the existing framework of outdoor noise modeling standards. Concerns related to documentation and to the protection of proprietary signal processing settings are resolved as well. The proposed approach was validated by measurements at a number of outdoor concerts. Results are shown to be practically accurate within the given limits of uncertainty.

This paper describes ongoing efforts toward optimizing the Virtual Acoustics Technology (VAT) system installed in the Immersive Media Lab at McGill University. Following the integration of the CAVIAR cancelling auralizer for feedback suppression, this current iteration of the active acoustics system is able to flexibly support the creation of virtual environments via the convolution of Spatial Room Impulse Responses (SRIRs) with real-time microphone signals. While the system has been successfully used for both recordings and live performances, we have nevertheless been looking to improve upon its “stability from feedback” and “natural sound quality,” two significant attributes of active acoustics systems [1]. We have implemented new software controls and microphone input methods to increase our ratio of gain before feedback, while additionally repositioning and adding loudspeakers to the system to generate a more even room coverage. Following these additions, we continue to evaluate the space through objective measurements and feedback from musicians and listeners.
[1] M. A. Poletti, “Active Acoustic Systems for the Control of Room Acoustics,” in Proceedings of the International Symposium on Room Acoustics, Melbourne, Australia, Aug. 2010.

The utilization of room impulse responses has proven valuable for both the acoustic assessment of indoor environments and music production. Various techniques have been devised over time to capture these responses. Although algorithmic solutions have been in existence since the 1960s for generating synthetic reverberation in real time, they continue to be computationally demanding and in general lack the accuracy in comparison to measured authentic Room Impulse Responses (RIR). In recent times, machine learning has found application in diverse fields, including acoustics, leading to the development of techniques for generating RIRs. This paper provides a general overview, of approaches and methods for generating RIRs, categorized into algorithmic and machine learning techniques, with a particular emphasis on the latter. Discussion covers the acoustical attributes of rooms relevant to perceptual testing and methodologies for comparing RIRs. An examination of disparities between captured and generated RIRs is included to better delineate the key acoustic properties characterizing a room. The paper is designed to offer a general overview for those interested in RIR generation for music production purposes, with future work considerations also explored.

This presentation provides an insightful overview of statistical acoustic calculation techniques and geometrical acoustic modeling. It addresses the unique challenges associated with small to medium-sized spaces and delves into the intricacies of contemporary ray tracing software tools. Attendees will also learn about the latest advancements of using these tools in immersive audio environments. Additionally, the presentation will showcase comparative low-frequency analysis calculations obtained using a variety of tools, including boundary element methods (BEM), finite element
methods (FEM) and others. The session includes a series of real-world case studies to illustrate these concepts in practical applications.

With consumers able to experience immersive audio more easily and cost-effectively than ever, and both Spotify as well as Apple Music platforms offering playout facilities for immersive content, demand is seemingly continuing to accelerate. The presentation offers a discussion of acoustical and architectural requirements and challenges as well as the relevant standards to adhere to. As a case study, the presentation will review Paris' newest studio, including the design strategies as well as advanced predictive acoustic analysis involved in creating the studio's design. It will also the
studio's unique elevator assisted “disappearing console” strategy for its Studio A Control Room changing into a multi-purpose live room and classroom.

Large scale systems face unique challenges of scale, shape, acoustic overflow restrictions, flown weight restrictions and yet still must produce high quality experiences for the audience. Join Bob McCarthy for a discussion of the large-scale outdoor systems used in festivals from the NY Philharmonic in Central Park through major rock festivals.

The majority of loudspeaker design now relies on simulation software such as Klippel. This approach has been effective in improving various speaker characteristics. However, the overreliance on simulation software without a deep understanding of electro-acoustic transducers has led to a neglect of the fundamental principles of speakers. This has become a bottleneck in enhancing the sound quality of speakers.

This session re-focuses on impedance characteristics, which are no longer receiving sufficient attention today. It explains the sound quality issues associated with impedance fluctuations, methods for addressing them, and how to design speakers based on information derived from impedance characteristics.