Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices

Poul Hoang, Jesper Jensen, Jan Mark de Haan, Zheng-Hua Tan, Thomas Lunner

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearch

Abstract

Multi-microphone speech enhancement systems often apply beamforming to enhance one or multiple desired signals in a noisy environment. Common for many beamforming methods, is that they require the direction-of-arrival (DOA) of the target sound source to be known in order to achieve optimal noise reduction performance. To improve robustness against DOA uncertainty, we propose maximum a posteriori (MAP) and Bayesian beamformers that are able to take advantage of prior information on the target direction. We compare the proposed MAP and Bayesian beamformers to state-of-the-art beamforming methods for noise reduction in hearing assistive devices. We evaluate the proposed beamformers in isotropic babble noise in terms of segmental SNR (SSNR) and extended short-time objective intelligibility (ESTOI). Results show that the proposed methods outperform current state-of-the-art beamformers used for noise reduction in hearing aids in most scenarios.
Original languageDanish
Title of host publicationIEEE Global Conference on Signal and Information Processing (GlobalSIP)
Publication statusAccepted/In press - 2020
SeriesIEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings

Cite this

Hoang, P., Jensen, J., de Haan, J. M., Tan, Z-H., & Lunner, T. (Accepted/In press). Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices. In IEEE Global Conference on Signal and Information Processing (GlobalSIP) IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings
Hoang, Poul ; Jensen, Jesper ; de Haan, Jan Mark ; Tan, Zheng-Hua ; Lunner, Thomas. / Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices. IEEE Global Conference on Signal and Information Processing (GlobalSIP). 2020. (IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings).
@inproceedings{81350f245a64492ab4da64f8e0b744d3,
title = "Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices",
abstract = "Multi-microphone speech enhancement systems often apply beamforming to enhance one or multiple desired signals in a noisy environment. Common for many beamforming methods, is that they require the direction-of-arrival (DOA) of the target sound source to be known in order to achieve optimal noise reduction performance. To improve robustness against DOA uncertainty, we propose maximum a posteriori (MAP) and Bayesian beamformers that are able to take advantage of prior information on the target direction. We compare the proposed MAP and Bayesian beamformers to state-of-the-art beamforming methods for noise reduction in hearing assistive devices. We evaluate the proposed beamformers in isotropic babble noise in terms of segmental SNR (SSNR) and extended short-time objective intelligibility (ESTOI). Results show that the proposed methods outperform current state-of-the-art beamformers used for noise reduction in hearing aids in most scenarios.",
keywords = "Beamforming, Noise Reduction, Hearing Aids, Bayesian Beamforming, Maximum a posterior beamforming",
author = "Poul Hoang and Jesper Jensen and {de Haan}, {Jan Mark} and Zheng-Hua Tan and Thomas Lunner",
year = "2020",
language = "Dansk",
series = "IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings",
publisher = "IEEE Signal Processing Society",
booktitle = "IEEE Global Conference on Signal and Information Processing (GlobalSIP)",

}

Hoang, P, Jensen, J, de Haan, JM, Tan, Z-H & Lunner, T 2020, Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices. in IEEE Global Conference on Signal and Information Processing (GlobalSIP). IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings.

Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices. / Hoang, Poul; Jensen, Jesper; de Haan, Jan Mark; Tan, Zheng-Hua; Lunner, Thomas.

IEEE Global Conference on Signal and Information Processing (GlobalSIP). 2020. (IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings).

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearch

TY - GEN

T1 - Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices

AU - Hoang, Poul

AU - Jensen, Jesper

AU - de Haan, Jan Mark

AU - Tan, Zheng-Hua

AU - Lunner, Thomas

PY - 2020

Y1 - 2020

N2 - Multi-microphone speech enhancement systems often apply beamforming to enhance one or multiple desired signals in a noisy environment. Common for many beamforming methods, is that they require the direction-of-arrival (DOA) of the target sound source to be known in order to achieve optimal noise reduction performance. To improve robustness against DOA uncertainty, we propose maximum a posteriori (MAP) and Bayesian beamformers that are able to take advantage of prior information on the target direction. We compare the proposed MAP and Bayesian beamformers to state-of-the-art beamforming methods for noise reduction in hearing assistive devices. We evaluate the proposed beamformers in isotropic babble noise in terms of segmental SNR (SSNR) and extended short-time objective intelligibility (ESTOI). Results show that the proposed methods outperform current state-of-the-art beamformers used for noise reduction in hearing aids in most scenarios.

AB - Multi-microphone speech enhancement systems often apply beamforming to enhance one or multiple desired signals in a noisy environment. Common for many beamforming methods, is that they require the direction-of-arrival (DOA) of the target sound source to be known in order to achieve optimal noise reduction performance. To improve robustness against DOA uncertainty, we propose maximum a posteriori (MAP) and Bayesian beamformers that are able to take advantage of prior information on the target direction. We compare the proposed MAP and Bayesian beamformers to state-of-the-art beamforming methods for noise reduction in hearing assistive devices. We evaluate the proposed beamformers in isotropic babble noise in terms of segmental SNR (SSNR) and extended short-time objective intelligibility (ESTOI). Results show that the proposed methods outperform current state-of-the-art beamformers used for noise reduction in hearing aids in most scenarios.

KW - Beamforming

KW - Noise Reduction

KW - Hearing Aids

KW - Bayesian Beamforming

KW - Maximum a posterior beamforming

M3 - Konferenceartikel i proceeding

T3 - IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings

BT - IEEE Global Conference on Signal and Information Processing (GlobalSIP)

ER -

Hoang P, Jensen J, de Haan JM, Tan Z-H, Lunner T. Robust Bayesian and Maximum a Posteriori Beamforming for Hearing Assistive Devices. In IEEE Global Conference on Signal and Information Processing (GlobalSIP). 2020. (IEEE Global Conference on Signal and Information Processing (GlobalSIP). Proceedings).