EUVIMEDEUVIMED entdecken
Crossref

Exploring frequency-domain feature modeling for head-related transfer functions magnitude upsampling

Xingyu Chen iD, Hanwen Bi iD, Fei Ma iD, Sipei Zhao iD, Eva Cheng iD, Ian S Burnett iD

The Journal of the Acoustical Society of America · 01.11.2026 · 160 (5) · 3663-3674

DOI: 10.1121/10.0047189 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, online erschienen am 02.11.2026. Verfasst von Xingyu Chen und 5 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Audio, Acoustics and Vibration, Faculty of Engineering and Information Technology, University of Technology Sydney, Ultimo, New South Wale…. Die Arbeit nennt 58 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate upsampling of head-related transfer functions (HRTFs) from sparse measurements is crucial for personalized spatial audio rendering. Traditional interpolation methods, such as kernel-based weighting or basis function expansions, rely on measurements from a single subject and are limited by the spatial sampling theorem, resulting in significant performance degradation under sparse sampling. Recent learning-based methods alleviate this limitation by leveraging cross-subject information, yet most existing neural architectures primarily focus on modeling spatial relationships across directions, whereas spectral dependencies along the frequency dimension are often modeled implicitly or treated independently. However, HRTF magnitude responses exhibit strong local continuity and long-range structure in the frequency domain, which are not fully exploited. This work investigates frequency-domain feature modeling by examining how different architectural choices—ranging from per-frequency multilayer perceptrons to convolutional, dilated convolutional, and attention-based models—affect performance under varying sparsity levels, showing that explicit spectral modeling consistently improves reconstruction accuracy, particularly under severe sparsity. Motivated by this observation, a frequency-domain Conformer-based architecture is adopted to jointly capture local spectral continuity and long-range frequency correlations. Experimental results on the SONICOM and HUTUBS datasets demonstrate that the proposed method achieves state-of-the-art performance in terms of interaural level difference and log-spectral distortion.

Autor:innen und Institutionen

  • Centre for Audio, Acoustics and Vibration, Faculty of Engineering and Information Technology, University of Technology Sydney, Ultimo, New South Wales 2007, Australia.
  • Faculty of Information Technology, Monash University, Clayton, Victoria 3800, Australia.

ORCID-Profile: Xingyu Chen, Hanwen Bi, Fei Ma, Sipei Zhao, Eva Cheng, Ian S Burnett

Zitieren

APA

Chen, X., Bi, H., Ma, F., Zhao, S., Cheng, E., & Burnett, I. S. (2026). Exploring frequency-domain feature modeling for head-related transfer functions magnitude upsampling. The Journal of the Acoustical Society of America, 160(5), 3663-3674. https://doi.org/10.1121/10.0047189

Vancouver

Chen X, Bi H, Ma F, Zhao S, Cheng E, Burnett IS. Exploring frequency-domain feature modeling for head-related transfer functions magnitude upsampling. The Journal of the Acoustical Society of America. 2026;160(5):3663-3674. doi:10.1121/10.0047189

BibTeX
@article{Chen2026,
  author = {Chen, Xingyu and Bi, Hanwen and Ma, Fei and Zhao, Sipei and Cheng, Eva and Burnett, Ian S},
  title = {Exploring frequency-domain feature modeling for head-related transfer functions magnitude upsampling},
  journal = {The Journal of the Acoustical Society of America},
  year = {2026},
  volume = {160},
  number = {5},
  pages = {3663--3674},
  doi = {10.1121/10.0047189}
}
Kostenlos starten →