HRTF personalization modeling based on RBF neural network

Li, Lin; Huang, Qinghua

doi:10.1109/icassp.2013.6638350

Cited by 28 publications

(16 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Note that when we concatenate HRTF values of all the HRTF directions into one dimensional data tensor, (4) is equivalent to (3). The perceptual meaning of LSD is equally unclear.…”

Section: Hrtf Metricsmentioning

confidence: 97%

See 1 more Smart Citation

HRTF magnitude synthesis via sparse representation of anthropometric features

Biliński

Ahrens²,

Thomas

et al. 2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

We propose a method for the synthesis of the magnitudes of Head-related Transfer Functions (HRTFs) using a sparse representation of anthropometric features. Our approach treats the HRTF synthesis problem as finding a sparse representation of the subject's anthropometric features w.r.t. the anthropometric features in the training set. The fundamental assumption is that the magnitudes of a given HRTF set can be described by the same sparse combination as the anthropometric data. Thus, we learn a sparse vector that represents the subject's anthropometric features as a linear superposition of the anthropometric features of a small subset of subjects from the training data. Then, we apply the same sparse vector directly on the HRTF tensor data. For evaluation purpose we use a new dataset, containing both anthropometric features and HRTFs. We compare the proposed sparse representation based approach with ridge regression and with the data of a manikin (which was designed based on average anthropometric data), and we simulate the best and the worst possible classifiers to select one of the HRTFs from the dataset. For instrumental evaluation we use log-spectral distortion. Experiments show that our sparse representation outperforms all other evaluated techniques, and that the synthesized HRTFs are almost as good as the best possible HRTF classifier.

show abstract

“…Note that when we concatenate HRTF values of all the HRTF directions into one dimensional data tensor, (4) is equivalent to (3). The perceptual meaning of LSD is equally unclear.…”

Section: Hrtf Metricsmentioning

confidence: 97%

“…Many techniques have been recently proposed for HRTF personalization [2][3][4][5][6][7][8][9][10][11] based on a selected set of anthropometric features. Their effectiveness heavily depends on the choice of anthropometric features.…”

Section: Introductionmentioning

confidence: 99%

HRTF magnitude synthesis via sparse representation of anthropometric features

Biliński

Ahrens²,

Thomas

et al. 2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Due to the fact that HRTF individualization is strongly related to the anthropometry of a person, methods have been proposed for HRTF personalization by choosing a small set of anthropometry features with a pre-trained model [49][50][51][52][53][54]. The training was established based on a direct linear or nonlinear relationship between the anthropometric data and the HRTFs, where the first step is to reduce the HRTF data dimensionality.…”

Section: Individualized Hrtfmentioning

confidence: 99%

Surround by Sound: A Review of Spatial Audio Recording and Reproduction

et al. 2017

View full text Add to dashboard Cite

Abstract:In this article, a systematic overview of various recording and reproduction techniques for spatial audio is presented. While binaural recording and rendering is designed to resemble the human two-ear auditory system and reproduce sounds specifically for a listener's two ears, soundfield recording and reproduction using a large number of microphones and loudspeakers replicate an acoustic scene within a region. These two fundamentally different types of techniques are discussed in the paper. A recent popular area, multi-zone reproduction, is also briefly reviewed in the paper. The paper is concluded with a discussion of the current state of the field and open problems.

show abstract

“…To get rid of trivial anthropometric measurements and improve the performance of the HRTF estimation, Hu et al used partial least squares regression (PLSR) to model the linear relation [24]. Subsequently, to further describe the scattering of the incident sound by the physical structures, many researchers explored some non-linear multivariable statistical estimation methods to improve the performance of HRTF customization [25,[30][31][32]. In [25], a three-layer back-propagation artificial neural network (ANN) was used to HRTF personalization.…”

Section: Introductionmentioning

confidence: 99%

Modeling individual HRTF tensor using high-order partial least squares

Huang

2014

EURASIP J. Adv. Signal Process.

Self Cite

View full text Add to dashboard Cite

A tensor is used to describe head-related transfer functions (HRTFs) depending on frequencies, sound directions, and anthropometric parameters. It keeps the multi-dimensional structure of measured HRTFs. To construct a multi-linear HRTF personalization model, an individual core tensor is extracted from the original HRTFs using high-order singular value decomposition (HOSVD). The individual core tensor in lower-dimensional space acts as the output of the multi-linear model. Some key anthropometric parameters as the inputs of the model are selected by Laplacian scores and correlation analyses between all the measured parameters and the individual core tensor. Then, the multi-linear regression model is constructed by high-order partial least squares (HOPLS), aiming to seek a joint subspace approximation for both the selected parameters and the individual core tensor. The numbers of latent variables and loadings are used to control the complexity of the model and prevent overfitting feasibly. Compared with the partial least squares regression (PLSR) method, objective simulations demonstrate the better performance for predicting individual HRTFs especially for the sound directions ipsilateral to the concerned ear. The subjective listening tests show that the predicted individual HRTFs are approximate to the measured HRTFs for the sound localization.

show abstract

HRTF personalization modeling based on RBF neural network

Cited by 28 publications

References 8 publications

HRTF magnitude synthesis via sparse representation of anthropometric features

HRTF magnitude synthesis via sparse representation of anthropometric features

Surround by Sound: A Review of Spatial Audio Recording and Reproduction

Modeling individual HRTF tensor using high-order partial least squares

Contact Info

Product

Resources

About