On Data-Driven Approaches to Head-Related-Transfer Function Personalization

Fayek H, van der Maaten L, Romigh G, Mehra R. On Data-Driven Approaches to Head-Related-Transfer Function Personalization. In: AES Convention 143. Audio Engineering Society; 2017. Paper 9890. Available from: https://aes.org/publications/elibrary-page/?id=19287

@inproceedings{Fayek2017_19287,
  author = {Fayek, Haytham and van der Maaten, Laurens and Romigh, Griffin and Mehra, Ravish},
  title = {{On Data-Driven Approaches to Head-Related-Transfer Function Personalization}},
  booktitle = {AES Convention 143},
  note = {Paper 9890},
  year = {2017},
  month = oct,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=19287}
}

TY  - CPAPER
TI  - On Data-Driven Approaches to Head-Related-Transfer Function Personalization
AU  - Fayek, Haytham
AU  - van der Maaten, Laurens
AU  - Romigh, Griffin
AU  - Mehra, Ravish
T2  - AES Convention 143
M1  - Paper 9890
PY  - 2017
DA  - 2017/10/06
UR  - https://aes.org/publications/elibrary-page/?id=19287
PB  - Audio Engineering Society
LA  - en
AB  - Head-Related Transfer Function (HRTF) personalization is key to improving spatial audio perception and localization in virtual auditory displays. We investigate the task of personalizing HRTFs from anthropometric measurements, which can be decomposed into two sub tasks: Interaural Time Delay (ITD) prediction and HRTF magnitude spectrum prediction. We explore both problems using state-of-the-art Machine Learning (ML) techniques. First, we show that ITD prediction can be significantly improved by smoothing the ITD using a spherical harmonics representation. Second, our results indicate that prior unsupervised dimensionality reduction-based approaches may be unsuitable for HRTF personalization. Last, we show that neural network models trained on the full HRTF representation improve HRTF prediction compared to prior methods.
ER  -