Learning-based multi-listener sound field reproduction using boundary velocity targets
Learning-based multi-listener sound field reproduction using boundary velocity targets
This paper studies learning-based sound field control for a prototype large-audience scenario with ten loudspeakers and multiple listeners distributed over a listening region. Two identical complex-valued multilayer perceptrons are compared. One is trained on a boundary normal-velocity objective and one is trained directly on binaural ear pressures. The velocity-based network is first validated against the analytical controller based on previous work by Shin et al. and shown to closely reproduce its region-based behavior. The two models are then compared using common region-based and listener-based metrics, including pressure error and binaural cue errors based on Interaural Level Difference (ILD) and Interaural Phase Difference (IPD). For ten listeners, direct ear pressure training defines an overdetermined 20-ear/10-loudspeaker problem, whereas the velocity-based model provides a more robust trade-off between regional accuracy and perceptual performance. The results indicate that region-based velocity training is a promising approach for overdetermined multi-listener sound field reproduction.
Paul, Vlad-Stefan
3f46807b-cda2-4818-ba70-71bfd26a1a92
Hahn, Nara
9c5cb8ff-b351-40ff-974b-9635a790ec16
Nelson, Philip
5c6f5cc9-ea52-4fe2-9edf-05d696b0c1a9
2026
Paul, Vlad-Stefan
3f46807b-cda2-4818-ba70-71bfd26a1a92
Hahn, Nara
9c5cb8ff-b351-40ff-974b-9635a790ec16
Nelson, Philip
5c6f5cc9-ea52-4fe2-9edf-05d696b0c1a9
Paul, Vlad-Stefan, Hahn, Nara and Nelson, Philip
(2026)
Learning-based multi-listener sound field reproduction using boundary velocity targets.
Forum Acusticum 2026, Messecongress Graz (MCG), Graz, Austria.
08 - 12 Sep 2026.
Record type:
Conference or Workshop Item
(Paper)
Abstract
This paper studies learning-based sound field control for a prototype large-audience scenario with ten loudspeakers and multiple listeners distributed over a listening region. Two identical complex-valued multilayer perceptrons are compared. One is trained on a boundary normal-velocity objective and one is trained directly on binaural ear pressures. The velocity-based network is first validated against the analytical controller based on previous work by Shin et al. and shown to closely reproduce its region-based behavior. The two models are then compared using common region-based and listener-based metrics, including pressure error and binaural cue errors based on Interaural Level Difference (ILD) and Interaural Phase Difference (IPD). For ten listeners, direct ear pressure training defines an overdetermined 20-ear/10-loudspeaker problem, whereas the velocity-based model provides a more robust trade-off between regional accuracy and perceptual performance. The results indicate that region-based velocity training is a promising approach for overdetermined multi-listener sound field reproduction.
This record has no associated files available for download.
More information
Published date: 2026
Venue - Dates:
Forum Acusticum 2026, Messecongress Graz (MCG), Graz, Austria, 2026-09-08 - 2026-09-12
Identifiers
Local EPrints ID: 515468
URI: https://eprints.soton.ac.uk/id/eprint/515468
PURE UUID: db3d8500-fea2-43b9-9c7f-fb032cfcdec9
Catalogue record
Date deposited: 14 Oct 2026 17:40
Last modified: 15 Oct 2026 03:00
Export record
Contributors
Author:
Vlad-Stefan Paul
Author:
Nara Hahn
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics