A Parallel Approach to HRTF Approximation and Interpolation Based on a Parametric Filter Model

Four head-related transfer function (HRTF) data sets were compared to determine the effect of HRTF measurement methodology on the localization of spatialized auditory stimuli. Spatial audio interfaces typically require HRTF data sets to generate the spatialized auditory stimuli. HRTF measurement is accomplished using a variety of techniques that can require several nearly arbitrary decisions about methodology. The effects of these choices upon the resulting spatial audio interface are unclear. Sixteen participants completed a sound localization task that included real-world, broadband stimuli spatialized at eight locations on the horizontal plane. Four different HRTF data sets were utilized to spatialize the stimuli: two publicly available HRTF data sets and two data sets obtained using different in-house measurement systems. All HRTFs were obtained from the Knowles Electronics Mannequin for Acoustics Research. Unsigned localization error and proportion of front/back reversals did not differ significantly across HRTF data sets. Poorest accuracy was observed in locations near the medial (front/back) axis of the listener, mainly because of the relatively large proportions of reversals at these locations. This study suggests that the particular generalized HRTF data set chosen for spatialization is of minimal importance to the localizability of the resulting stimuli. This result will inform the design of many spatial audio interfaces that are based upon generalized HRTFs, including wayfaring devices, communication systems, and virtual reality systems.

The extended use of mobile multimedia devices in applications like gaming, 3D video and audio reproduction, immersive teleconferencing, or virtual and augmented reality, is demanding efficient algorithms and methodologies. All these applications require real-time spatial audio engines with the capability of dealing with intensive signal processing operations while facing a number of constraints related to computational cost, latency and energy consumption. Most mobile multimedia devices include a Graphics Processing Unit (GPU) that is primarily used to accelerate video processing tasks, providing high computational capabilities due to its inherent parallel architecture. This paper describes a scalable parallel implementation of a real-time binaural audio engine for GPU-equipped mobile devices. The engine is based on a set of head-related transfer functions (HRTFs) modelled with a parametric parallel structure, allowing efficient synthesis and interpolation while reducing the size required for HRTF data storage. Several strategies to optimize the GPU implementation are evaluated over a well-known kind of processor present in a wide range of mobile devices. In this context, we analyze both the energy consumption and real-time capabilities of the system by exploring different GPU and CPU configuration alternatives. Moreover, the implementation has been conducted using the OpenCL framework, guarantying the portability of the code.

Head-Related Transfer Function (HRTF) plays a critical role in how the auditory system perceives spatial information. The spectral cues embedded in HRTF are vital for accurately determining the elevation of sound sources. In existing approaches, deep neural networks (DNNs) have been utilized to predict the magnitude spectra of HRTF from images of the pinna, typically employing the HRTF log-magnitude as the output during training. However, HRTF encompasses the acoustic characteristics of both the head and torso, exhibiting direction-dependent patterns that pose challenges in reconstructing its spectral cues. To address this complexity, we propose an innovative method for HRTF individualization. Our model uses Pinna-Related Transfer Function (PRTF) as the output during training, which helps alleviate the impact of sound reflections from the head and torso in the head-related impulse response (HRIR). Our experimental findings, based on an HRTF dataset, illustrate that our proposed model excels in reconstructing the first and second spectral cues. Furthermore, it outperforms previous deep learning models in terms of log spectral distortion (LSD).

Head-related transfer function (HRTF) is characterized as sound transmission from sound source to listener's eardrum. When a listener hears a sound that is filtered with the HRTFs, the listener can localize a virtual target (sound image) as if the sound had come from the position corresponding to that at which the HRTFs were measured. A moving sound image can be generated to switch HRTFs of successive direction in real-time. While many virtual auditory displays (VADs) based on synthesis of HRTFs have been proposed, most of them can synthesize only a few sound images due to lack of computation power. In this article, the VAD system implemented based on graphics processing unit (GPU) was introduced. In our system, the convolution of HRTFs is parallelized on GPU to realize a high-speed processing. In addition, the multiple HRTFs each of which is corresponding to sound sources at different position are processed in parallel to control multiple sound image simultaneously. In this article, the performance of our system was evaluated not only objectively but also subjectively. The results showed that our current system can present at least 40 sound images simultaneously in real-time.

This paper introduces a continuous functional model for head-related transfer functions (HRTFs) in the horizontal auditory scene. The approach uses a separable representation consisting of a Fourier-Bessel series expansion for the spectral components and a conventional Fourier series expansion for the spatial components. Being independent of the data, these two sets of basis functions remain unchanged for all subjects and measurement setups. Hence, the model can transform an individualized HRTF to a subject specific set of coefficients. A continuous functional model is also developed in the time domain. We show the efficient model performance in approximating experimental measurements by using the HRTF measurements from a KEMAR manikin and the synthetic data from the spherical head model. The statistical results are determined from a 50-subject HRTF data set. We also corroborate the predictive capability of the proposed model. The model has near optimal performance, which can be ascertained by comparison with the standard principle component analysis (PCA) and discrete Karhunen-Loeve expansion (KLE) methods at the measurement points and for a given number of parameters.

이 연구는 중앙면 상에서 주성분 분석법을 이용하여 시간 및 주파수 영역에서 머리전달함수의 모형화 기법들을 다룬다. 시간영역의 머리전달함수, 복소수 값의 머리전달함수, 확장된 머리전달함수, 로그 크기의 머리전달함수에 기반하여 각각 주성분 분석법을 수행하여 얻은 네 가지 머리전달함수 모형들에 대해서 최소자승오차 관점에서 모형화 성능을 비교하고, 모형들간의 이론적인 관계를 살펴보는 것이 이 연구의 목적이다. 모형화에 사용되는 기저함수의 수가 동일하다면, 시간영역에서의 머리전달함수 혹은 확장된 머리전달함수에 기반한 모형이 복소수 값의 머리전달함수에 기반한 모형보다 최소자승오차 관점에서 더 효율적인 모형화 성능을 지닌다. 시간영역에서의 머리전달함수에 기반한 모형과 확장된 머리전달함수에 기반한 모형은 이론적으로 동일한 모형이며 서로 푸리에 변환 관계가 있다. 로그 크기의 머리전달함수에 기반한 모형은 다른 모형들과 모형화 성능 및 이론적인 관계를 비교할 수가 없는데, 이는 로그 크기의 머리전달함수에 기반한 모형은 머리전달함수의 크기 정보만을 로그 크기로 다루는 반면에 다른 모형들은 선형 크기로 머리전달함수의 크기와 위상정보를 모두 다루기 때문이다. This study deals with modeling of head-related transfer functions(HRTFs) using principal components analysis(PCA) in the time and frequency domains. Four PCA models based on head-related impulse responses(HRIRs), complex-valued HRTFs, augmented HRTFs, and log-magnitudes of HRTFs are investigated. The objective of this study is to compare modeling performances of the PCA models in the least-squares sense and to show the theoretical relationship between the PCA models. In terms of the number of principal components needed for modeling, the PCA model based on HRIR or augmented HRTFs showed more efficient modeling performance than the PCA model based on complex-valued HRTFs. The PCA model based on HRIRs in the time domain and that based on augmented HRTFs in the frequency domain are shown to be theoretically equivalent. Modeling performance of the PCA model based on log-magnitudes of HRTFs cannot be compared with that of other PCA models because the PCA model deals with log-scaled magnitude components only, whereas the other PCA models consider both magnitude and phase components in linear scale.

The head-related transfer function (HRTF) defines the acoustic path from a source to the two ears of a listener in a manner that is highly dependent on direction. This directional dependence arises from the highly individual morphology of the pinna, which results in complex reflections and resonances. While this notion is generally accepted, there has been little research on the importance of different structural elements of the pinna on the HRTF. A parametric three-dimensional ear model was used to investigate the changes in shape of the pinna in a systematic manner with a view to determining important contributing morphological parameters that can be used for HRTF individualization. HRTFs were simulated using the boundary element method. The analysis comprised objective comparisons between the directional transfer function and diffuse field component. The mean spectral distortion was used for global evaluation of HRTF similarity across all simulated positions. A perceptual localization model was used to determine correspondences between perceptual cues and objective parameters. A reasonable match was found between the modelled perceptual results and the mean spectral distortion. Modifications to the shape of the concha were found to have an important impact on the HRTF, as did those in proximity to the triangular fossa. Furthermore, parameters that control the relief of the pinna were found to be at least as important as more frequently cited side-facing parameters, highlighting limitations in previous morphological/HRTF studies.

Head-related transfer function (HRTF) plays an important role in the construction of 3D auditory display. This paper presents an individual HRTF modeling method using deep neural networks based on spatial principal component analysis. The HRTFs are represented by a small set of spatial principal components combined with frequency and individual-dependent weights. By estimating the spatial principal components using deep neural networks and mapping the corresponding weights to a quantity of anthropometric parameters, we predict individual HRTFs in arbitrary spatial directions. The objective and subjective experiments evaluate the HRTFs generated by the proposed method, the principal component analysis (PCA) method, and the generic method. The results show that the HRTFs generated by the proposed method and PCA method perform better than the generic method. For most frequencies the spectral distortion of the proposed method is significantly smaller than the PCA method in the high frequencies but significantly larger in the low frequencies. The evaluation of the localization model shows the PCA method is better than the proposed method. The subjective localization experiments show that the PCA and the proposed methods have similar performances in most conditions. Both the objective and subjective experiments show that the proposed method can predict HRTFs in arbitrary spatial directions.

The KEMAR Head and Torso Simulator is one of the most commonly used artificial head and torso measurement systems and its head-related transfer function (HRTF) is often used in 3D-Audio applications. As part of this publication, an HRTF measurement performed in an hemi-anechoic chamber as well as a high resolution 3D-Model of the same is provided. This accompanying report details the data acquisition and post-processing steps, that resulted in the provided data sets including a discussion of the shortcomings of the available measurement setup.

Personalised binaural audio requires individual head-related transfer functions (HRTFs). Nowadays, it is feasible to compute HRTFs numerically from meshes of a listener’s anatomy. These meshes require geometric details of the pinna which are difficult to capture and often corrupted by noise and outliers. The alignment of a highresolution template geometry to the captured listener geometry seems to be a promising approach to deal with these corruptions. However, such an alignment represents a multi-dimensional optimisation problem. To tackle this, we describe a parametric pinna model which reduces the representation of a non-pathological human pinna geometry from a few thousand points in space to 144 parameter dimensions. The model was evaluated in the geometric and psychoacoustic domains in terms of the Hausdorff distance distribution and simulated localisation errors, respectively. When aligned with individual ground-truth meshes, the parametric pinna model was able to represent these meshes with a small geometric error. The predicted sound-localisation performance, using the aligned meshes as a basis for HRTF calculations, was similar to that of the corresponding listener. These results indicate that the proposed parametric pinna model represents a versatile tool to create human pinna meshes with a detailed geometry for the calculation of individual HRTFs.

This paper presents a computationally efficient auralization method, which combines a wavelet-based model for head related transfer functions (HRTFs) with the grouping of such functions for close directions. The proposed method results in a significant reduction of the processing time of the auralization process, by modeling the HRTFs using wavelet transforms and grouping them in regions around the listener, without compromising significantly the identification of a source sound direction. Simulation results obtained with the proposed auralization process are shown in the paper.

Head-Related Transfer Function (HRTF) measurement and extraction are important tasks for personalized-spatial audio. Many laboratories have their own apparatuses for data-collection but few studies have compared their results to a common subject or have modeled inter-dataset variances. We present a Bayesian fusion method based on Gaussian process (GP) modeling of joint spatial-frequency HRTFs over different spherical-measurement grids. Neumann KU-100 dummy HRTFs from 7 labs in the "Club Fritz" study are compared and fused to each other based on learning a set of transformations from the GP data-likelihood and covariance assumptions; parameter and hyperparameter training is automatic. Experimental results show that fused models for horizontal and median-plane HRTFs generalize the datasets better than pre-transformed ones.

Human beings have a remarkable ability to determine the direction of arrival of a sound and to separate sounds of interest. Replicating this ability is a challenging problem in audio signal processing. In this paper we present a model for the head related transfer function (HRTF) developed with the localisation objective in mind. This is achieved by splitting the 3D localisation cues in terms of two functions which can be independently evaluated. We illustrate the theory for calculating these functions and validate the results against actual HRTF data. We find the model to be a close match for a significant number of potential source locations.

Purpose – The purpose of this paper is to present a system for customized binaural audio delivery based on the extraction of relevant features from a 2-D representation of the listener’s pinna. Design/methodology/approach – The most significant pinna contours are extracted by means of multi-flash imaging, and they provide values for the parameters of a structural head-related transfer function (HRTF) model. The HRTF model spatializes a given sound file according to the listener’s head orientation, tracked by sensor-equipped headphones, with respect to the virtual sound source. Findings – A preliminary localization test shows that the model is able to statically render the elevation of a virtual sound source better than non-individual HRTFs. Research limitations/implications – Results encourage a deeper analysis of the psychoacoustic impact that the individualized HRTF model has on perceived elevation of virtual sound sources. Practical implications – The model has low complexity and is suitable for implementation on mobile devices. The resulting hardware/software package will hopefully allow an easy and low-tech fruition of custom spatial audio to any user. Originality/value – The authors show that custom binaural audio can be successfully deployed without the need of cumbersome subjective measurements.

We consider immersive rendering of a set of sound sources, apparently positioned at different locations in three-dimensional (3-D) space around a listener. Each sound source is associated with a subband-coded audio stream. A low-complexity rendering approach, based on reduced-rank models of subband-domain Head-related transfer functions (HRTFs), is proposed in this paper. HRTFs are approximated using a set of basis vectors in subband-domain. Immersive rendering of partially decoded audio stream is performed in subband-domain using the basis vectors. Perceptual objective error measurements and subjective test results indicate, despite low computational complexity, this technique is effective in rendering of coded audio streams in 3-D space.