Spatial Perception of Sound Source Distribution in the Median Plane

Pulkki, Ville; Pöntynen, Henri; Santala, Olli

doi:10.17743/jaes.2019.0033

Cited by 23 publications

(32 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ensemble width was restricted to fixed azimuth limits of ± 30°during the convolution procedure (this limit was adopted for consistency with our previous studies [7,9]), with the angles randomly selected for individual sources. Despite the relatively narrow intended ensemble width, informal listening tests undertaken by these authors showed that the perceived width for some stimuli was much wider than intended, spanning almost all semicircles (± 90°), which is consistent with the observation made by Pulkki et al [31]. HRTFs were not interpolated during the convolution procedure.…”

Section: Repository Of Binaural Music Recordingssupporting

confidence: 83%

Automatic discrimination between front and back ensemble locations in HRTF-convolved binaural recordings of music

Zieliński

Antoniuk

Johnson

2022

J AUDIO SPEECH MUSIC PROC.

View full text Add to dashboard Cite

One of the greatest challenges in the development of binaural machine audition systems is the disambiguation between front and back audio sources, particularly in complex spatial audio scenes. The goal of this work was to develop a method for discriminating between front and back located ensembles in binaural recordings of music. To this end, 22, 496 binaural excerpts, representing either front or back located ensembles, were synthesized by convolving multi-track music recordings with 74 sets of head-related transfer functions (HRTF). The discrimination method was developed based on the traditional approach, involving hand-engineering of features, as well as using a deep learning technique incorporating the convolutional neural network (CNN). According to the results obtained under HRTF-dependent test conditions, CNN showed a very high discrimination accuracy (99.4%), slightly outperforming the traditional method. However, under the HRTF-independent test scenario, CNN performed worse than the traditional algorithm, highlighting the importance of testing the algorithms under HRTF-independent conditions and indicating that the traditional method might be more generalizable than CNN. A minimum of 20 HRTFs are required to achieve a satisfactory generalization performance for the traditional algorithm and 30 HRTFs for CNN. The minimum duration of audio excerpts required by both the traditional and CNN-based methods was assessed as 3 s. Feature importance analysis, based on a gradient attribution mapping technique, revealed that for both the traditional and the deep learning methods, a frequency band between 5 and 6 kHz is particularly important in terms of the discrimination between front and back ensemble locations. Linear-frequency cepstral coefficients, interaural level differences, and audio bandwidth were identified as the key descriptors facilitating the discrimination process using the traditional approach.

show abstract

Section: Repository Of Binaural Music Recordingssupporting

confidence: 83%

Automatic discrimination between front and back ensemble locations in HRTF-convolved binaural recordings of music

Zieliński

Antoniuk

Johnson

2022

J AUDIO SPEECH MUSIC PROC.

View full text Add to dashboard Cite

show abstract

“…The reduction of the receiver to a virtual probe that does not maintain directional cues does not suffice a comprehensive auralisation output and must be expanded by a perception-equivalent model (e.g., a binaural processor, cf. Section 4.7.2) or an acoustic reproduction system that maintains perceptual cues (such as Vector-Base Amplitude Panning (VBAP) [Pul97], Higher-Order Ambisonics (HOA) [Ger83] or Wave Field Synthesis (WFS) [BVV93], cf. also Section 4.8).…”

Section: Iso 9613mentioning

confidence: 99%

Real-Time Auralisation of Outdoor Sound Propagation

Stienen¹

2023

View full text Add to dashboard Cite

Auralisation describes the process of generating and presenting audible sound using computer programs and audio hardware. Since the result is perceived naturally by the human's auditory system, a demonstration by means of auralisation is easily comprehensible and does neither require background knowledge nor expertise. Producing auralisation under real-time constraints increases the implementation demands significantly. Real-time auralisation is required in applications that respond to user interaction, for example, in interactive Virtual Reality (VR) environments. Dynamically moving sound sources and receivers evoke a change in the perceived sound, and the corresponding result must be provided as quickly as possible. A feeling of immersion can be created if the response time of the system does not exceed perceptual thresholds and thus enables a plausible scene presentation. To achieve this emotional state, real-time auralisation must comply with the expected physical behavior. Because auralisation has gained much attention in room acoustics, the established concepts and approaches are insufficient, if applied to outdoor scenarios. This is due to the fact that simulation principles based on Geometrical Acoustics (GA) can determine specular sound reflection with ease, but only rudimentarily incorporate the contribution of diffracted sound. Nonetheless, the transmission of sound from a source to a receiver in an urban setting contains a significant contribution of reflected and diffracted components. Furthermore, highly dynamic virtual environments are time-variant. Traditional realisations neglect this characteristic, treating an acoustic environment as temporarily static on a frame-by-frame basis. Adaption to dynamic events is implemented by update routines performing sequences of time-invariant simulations, which is contradictory to the highly dynamic nature of outdoor scenarios. This misconception is addressed in this dissertation and an alternative solution is suggested. The realisation of a real-time auralisation application for outdoor environments represents a promising addition to current noise assessment procedures. It also delivers the foundation for auditory modality in VR regarding outdoor scenarios, serving as an audio-video tool for scientific investigations of urban sound environments, accounting for perceptual aspects. der PDs und die Zuordnung individueller Wellenfronten basiert auf einer Winkeldistanzmetrik ("k-means clustering") und die Korrektur der ITD berücksichtigt die Abweichungen im Azimut. Die Machbarkeit des vorgeschlagenen Echtzeit-Auralisierungssystems und die Eignung für die Schallausbreitung im Freien werden in einer interaktiven Applikation demonstriert, die auf VR-Technologien basiert.For the young who are inspiring me, the old who ground me and my dearest who holds everything together.

show abstract

“…Using the warped VRS directions, the contributions of the reflection coefficients of each of the orthogonal shoebox walls are calculated using vector base amplitude panning (VBAP; [47]) with the inverted (outside-pointing) wall normals as vector base. Eq.…”

Section: Extension To Inhomogeneous Boundary Conditions For Shoebox G...mentioning

confidence: 99%

“…Eq. ( 18) from [47] is therefore adapted to yield the contribution g k = [g 1 g 2 g 3 ] of the absorption coefficients of the three intersecting walls in each octant with the normal vectors n 1 , n 2 , n 3 to the filtering applied to a particular VRS with index k:…”

Section: Extension To Inhomogeneous Boundary Conditions For Shoebox G...mentioning

confidence: 99%

Computationally-Efficient Simulation of Late Reverberation for Inhomogeneous Boundary Conditions and Coupled Rooms

Kirsch¹,

Wendt²,

Par³

et al. 2023

J. Audio Eng. Soc.

View full text Add to dashboard Cite

Spatial Perception of Sound Source Distribution in the Median Plane

Cited by 23 publications

References 12 publications

Automatic discrimination between front and back ensemble locations in HRTF-convolved binaural recordings of music

Automatic discrimination between front and back ensemble locations in HRTF-convolved binaural recordings of music

Real-Time Auralisation of Outdoor Sound Propagation

Computationally-Efficient Simulation of Late Reverberation for Inhomogeneous Boundary Conditions and Coupled Rooms

Contact Info

Product

Resources

About