Global HRTF Personalization Using Anthropometric Measures
×
Cite This
Citation & Abstract
Y. Wang, Y. Zhang, Z. Duan, and M. Bocko, "Global HRTF Personalization Using Anthropometric Measures," Paper 10502, (2021 May.). doi:
Y. Wang, Y. Zhang, Z. Duan, and M. Bocko, "Global HRTF Personalization Using Anthropometric Measures," Paper 10502, (2021 May.). doi:
Abstract: In this paper, we propose an approach for global HRTF personalization employing subjects’ anthropometric features using spherical harmonics transform (SHT) and convolutional neural network (CNN). Existing methods employ different models for each elevation, which fails to take advantage of the underlying common features of the full set of HRTF’s. Using the HUTUBS HRTF database as our training set, a SHT was used to produce subjects’ personalized HRTF’s for all spatial directions using a single model. The resulting predicted HRTFs have a log-spectral distortion (LSD) level of 3.81 dB in comparison to the SHT reconstructed HRTFs, and 4.74 dB in comparison to the measured HRTFs. The personalized HRTFs show significant improvement upon the finite element acoustic computations of HRTFs provided in the HUTUBS database.
@article{wang2021global,
author={wang, yuxiang and zhang, you and duan, zhiyao and bocko, mark},
journal={journal of the audio engineering society},
title={global hrtf personalization using anthropometric measures},
year={2021},
volume={},
number={},
pages={},
doi={},
month={may},}
@article{wang2021global,
author={wang, yuxiang and zhang, you and duan, zhiyao and bocko, mark},
journal={journal of the audio engineering society},
title={global hrtf personalization using anthropometric measures},
year={2021},
volume={},
number={},
pages={},
doi={},
month={may},
abstract={in this paper, we propose an approach for global hrtf personalization employing subjects’ anthropometric features using spherical harmonics transform (sht) and convolutional neural network (cnn). existing methods employ different models for each elevation, which fails to take advantage of the underlying common features of the full set of hrtf’s. using the hutubs hrtf database as our training set, a sht was used to produce subjects’ personalized hrtf’s for all spatial directions using a single model. the resulting predicted hrtfs have a log-spectral distortion (lsd) level of 3.81 db in comparison to the sht reconstructed hrtfs, and 4.74 db in comparison to the measured hrtfs. the personalized hrtfs show significant improvement upon the finite element acoustic computations of hrtfs provided in the hutubs database.},}
TY - paper
TI - Global HRTF Personalization Using Anthropometric Measures
SP -
EP -
AU - Wang, Yuxiang
AU - Zhang, You
AU - Duan, Zhiyao
AU - Bocko, Mark
PY - 2021
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - May 2021
TY - paper
TI - Global HRTF Personalization Using Anthropometric Measures
SP -
EP -
AU - Wang, Yuxiang
AU - Zhang, You
AU - Duan, Zhiyao
AU - Bocko, Mark
PY - 2021
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - May 2021
AB - In this paper, we propose an approach for global HRTF personalization employing subjects’ anthropometric features using spherical harmonics transform (SHT) and convolutional neural network (CNN). Existing methods employ different models for each elevation, which fails to take advantage of the underlying common features of the full set of HRTF’s. Using the HUTUBS HRTF database as our training set, a SHT was used to produce subjects’ personalized HRTF’s for all spatial directions using a single model. The resulting predicted HRTFs have a log-spectral distortion (LSD) level of 3.81 dB in comparison to the SHT reconstructed HRTFs, and 4.74 dB in comparison to the measured HRTFs. The personalized HRTFs show significant improvement upon the finite element acoustic computations of HRTFs provided in the HUTUBS database.
In this paper, we propose an approach for global HRTF personalization employing subjects’ anthropometric features using spherical harmonics transform (SHT) and convolutional neural network (CNN). Existing methods employ different models for each elevation, which fails to take advantage of the underlying common features of the full set of HRTF’s. Using the HUTUBS HRTF database as our training set, a SHT was used to produce subjects’ personalized HRTF’s for all spatial directions using a single model. The resulting predicted HRTFs have a log-spectral distortion (LSD) level of 3.81 dB in comparison to the SHT reconstructed HRTFs, and 4.74 dB in comparison to the measured HRTFs. The personalized HRTFs show significant improvement upon the finite element acoustic computations of HRTFs provided in the HUTUBS database.
Authors:
Wang, Yuxiang; Zhang, You; Duan, Zhiyao; Bocko, Mark
Affiliation:
University of Rochester, Rochester, NY, USA
AES Convention:
150 (May 2021)
Paper Number:
10502
Publication Date:
May 24, 2021Import into BibTeX
Subject:
HRTF
Permalink:
http://www.aes.org/e-lib/browse.cfm?elib=21095