Y. Li, L. Wang, and J. Reiss, “Neural Regularization for Personal Sound Zones,” in Proc. AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games, Jun. 2026, pp. 10346:1-9. [Online]. Available: https://aes.org/publications/elibrary-page/?id=23308
Li Y, Wang L, Reiss J. Neural Regularization for Personal Sound Zones. In: AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games. Audio Engineering Society; 2026.pp. 10346:1-9. Available from: https://aes.org/publications/elibrary-page/?id=23308
@inproceedings{Li2026_23308,
author = {Li, Yazhou and Wang, Lin and Reiss, Joshua},
title = {{Neural Regularization for Personal Sound Zones}},
booktitle = {AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games},
pages = {10346:1-9},
url = {https://aes.org/publications/elibrary-page/?id=23308},
year = {2026},
month = jun,
publisher = {Audio Engineering Society},
address = {Paris France}
}
TY - CPAPER
TI - Neural Regularization for Personal Sound Zones
AU - Li, Yazhou
AU - Wang, Lin
AU - Reiss, Joshua
T2 - AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games
SP - 1
EP - 9
M1 - Paper 10346
PY - 2026
DA - 2026/06/30
UR - https://aes.org/publications/elibrary-page/?id=23308
PB - Audio Engineering Society
CY - Paris France
LA - en
AB - Pressure-matching (PM) for personal sound zone (PSZ) can achieve high contrast at nominal control points, but the performance may degrade when transfer functions are mismatched. We introduce a neural method that maps transfer functions to loudspeaker weights using a single-frequency input network with parameters shared across frequencies. We evaluate the robustness under position shifts, additive transfer-function noise, and added reflections, and compare against PM with Tikhonov regularization. Results show improved robustness to structured perturbations such as listener displacement, whereas regularized PM remains more resilient to unstructured random transfer-function noise and reverberation. We further explain these results using a singular value decomposition based perturbation projection. Finally, we analyze different regularization mechanisms induced by the network and derive practical guidelines for neural PSZ filter optimization.
ER -