Personalized HRTF Modeling Using DNN-Augmented BEM

Mengfan Zhang; Jui-Hsien Wang; Doug L. James

ICASSP 2021accepted0 citations

Personalized HRTF Modeling Using DNN-Augmented BEM

Mengfan Zhang, Jui-Hsien Wang, Doug L. James

Abstract

Accurate modeling of personalized head-related transfer functions (HRTFs) is difficult but critical for applications requiring spatial audio. However, this remains challenging as experimental measurements require specialized equipment, numerical simulations require accurate head geometries and robust solvers, and data-driven methods are hungry for data. In this paper, we propose a new deep learning method that combines measurements and numerical simulations to take the best of three worlds. By learning the residual difference and establishing a high quality spatial basis, our method achieves consistently 2 dB to 2.5 dB lower spectral distortion (SD) compared to the state-of-the-art methods.

BibTeX

@inproceedings{icassp2021_personalizedhrtf,
  title = {Personalized HRTF Modeling Using DNN-Augmented BEM},
  author = {Mengfan Zhang and Jui-Hsien Wang and Doug L. James},
  booktitle = {ICASSP 2021},
  year = {2021}
}