Time-frequency-bin-wise beamformer selection and masking for speech enhancement in underdetermined noisy scenarios

Yamaoka K, Brendel A, Ono N, Makino S, Bürger M, Yamada T, Kellermann W (2018)

Publication Language: English

Publication Type: Conference contribution

Publication year: 2018

Pages Range: 1596-1600

Event location: Rome IT

ISBN: 978-90-827970-1-5

DOI: 10.23919/eusipco.2018.8553299

Abstract

In this paper, we present a speech enhancement method using two microphones for underdetermined situations. A conventional speech enhancement method for underdetermined situations is time-frequency masking, where speech is enhanced by multiplying zero or one to each time-frequency component appropriately. Extending this method, we switch multiple preconstructed beamformers at each time-frequency bin, each of which suppresses a particular interferer. This method can suppress an interferer even when both the target and an interferer are simultaneously active at a given time-frequency bin. As a switching criterion, selection of minimum value of the outputs of the all beamformers at each time-frequency bin is investigated. Additionally, another method using direction of arrival estimation is also investigated. In experiments, we confirmed that the proposed methods were superior to conventional time-frequency masking and fixed beamforming in the performance of speech enhancement.

Authors with CRIS profile

Andreas Brendel Michael Bürger Professur für Signalverarbeitung Walter Kellermann Professur für Signalverarbeitung

Involved external institutions

Tokyo Metropolitan University (TMU) / 首都大学東京 JP Japan (JP) University of Tsukuba / 筑波大学 JP Japan (JP) Institute of Electrical and Electronics Engineers (IEEE) US United States (USA) (US)

How to cite

APA:

Yamaoka, K., Brendel, A., Ono, N., Makino, S., Bürger, M., Yamada, T., & Kellermann, W. (2018). Time-frequency-bin-wise beamformer selection and masking for speech enhancement in underdetermined noisy scenarios. In Proceedings of the European Signal Processing Conference (EUSIPCO) (pp. 1596-1600). Rome, IT.

MLA:

Yamaoka, Kouei, et al. "Time-frequency-bin-wise beamformer selection and masking for speech enhancement in underdetermined noisy scenarios." Proceedings of the European Signal Processing Conference (EUSIPCO), Rome 2018. 1596-1600.

BibTeX: Download