Multi-channel PSD Estimators for Speech Dereverberation: a Theoretical and Experimental Comparison

Adam Kuklasinski, Simon Doclo, Timo Gerkmann, Søren Holdt Jensen, Jesper Jensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

12 Citations (Scopus)

Abstract

In this paper we perform an extensive theoretical and experimental comparison of two recently proposed multi-channel speech dereverberation algorithms. Both of them are based on the multi-channel Wiener filter but they use different estimators of the speech and reverberation power spectral densities (PSDs). We first derive closed-form expressions for the mean square error (MSE) of both PSD estimators and then show that one estimator – previously used for speech dereverberation by the authors – always yields a better MSE. Only in the case of a two microphone array or for special spatial distributions of the interference both estimators yield the same MSE. The theoretically derived MSE values are in good agreement with numerical simulation results and with instrumental speech quality measures in a realistic speech dereverberation task for binaural hearing aids.
Original languageEnglish
Title of host publicationIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015
Number of pages5
PublisherIEEE Signal Processing Society
Publication date19 Apr 2015
Pages91-95
DOIs
Publication statusPublished - 19 Apr 2015
Event40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015 - Brisbane, Australia
Duration: 19 Apr 201524 Apr 2015
Conference number: 2015

Conference

Conference40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015
Number2015
CountryAustralia
CityBrisbane
Period19/04/201524/04/2015
SeriesI E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings
ISSN1520-6149

Fingerprint

estimators
binaural hearing
reverberation
microphones
spatial distribution
interference
filters
simulation

Cite this

Kuklasinski, A., Doclo, S., Gerkmann, T., Jensen, S. H., & Jensen, J. (2015). Multi-channel PSD Estimators for Speech Dereverberation: a Theoretical and Experimental Comparison. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015 (pp. 91-95). IEEE Signal Processing Society. I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings https://doi.org/10.1109/ICASSP.2015.7177938
Kuklasinski, Adam ; Doclo, Simon ; Gerkmann, Timo ; Jensen, Søren Holdt ; Jensen, Jesper. / Multi-channel PSD Estimators for Speech Dereverberation : a Theoretical and Experimental Comparison. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015. IEEE Signal Processing Society, 2015. pp. 91-95 (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings).
@inproceedings{64449c54cb3e4adda4ab4c85cb5518e5,
title = "Multi-channel PSD Estimators for Speech Dereverberation: a Theoretical and Experimental Comparison",
abstract = "In this paper we perform an extensive theoretical and experimental comparison of two recently proposed multi-channel speech dereverberation algorithms. Both of them are based on the multi-channel Wiener filter but they use different estimators of the speech and reverberation power spectral densities (PSDs). We first derive closed-form expressions for the mean square error (MSE) of both PSD estimators and then show that one estimator – previously used for speech dereverberation by the authors – always yields a better MSE. Only in the case of a two microphone array or for special spatial distributions of the interference both estimators yield the same MSE. The theoretically derived MSE values are in good agreement with numerical simulation results and with instrumental speech quality measures in a realistic speech dereverberation task for binaural hearing aids.",
author = "Adam Kuklasinski and Simon Doclo and Timo Gerkmann and Jensen, {S{\o}ren Holdt} and Jesper Jensen",
year = "2015",
month = "4",
day = "19",
doi = "10.1109/ICASSP.2015.7177938",
language = "English",
pages = "91--95",
booktitle = "IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015",
publisher = "IEEE Signal Processing Society",
address = "United States",

}

Kuklasinski, A, Doclo, S, Gerkmann, T, Jensen, SH & Jensen, J 2015, Multi-channel PSD Estimators for Speech Dereverberation: a Theoretical and Experimental Comparison. in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015. IEEE Signal Processing Society, I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings, pp. 91-95, 40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015, Brisbane, Australia, 19/04/2015. https://doi.org/10.1109/ICASSP.2015.7177938

Multi-channel PSD Estimators for Speech Dereverberation : a Theoretical and Experimental Comparison. / Kuklasinski, Adam; Doclo, Simon; Gerkmann, Timo; Jensen, Søren Holdt; Jensen, Jesper.

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015. IEEE Signal Processing Society, 2015. p. 91-95.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Multi-channel PSD Estimators for Speech Dereverberation

T2 - a Theoretical and Experimental Comparison

AU - Kuklasinski, Adam

AU - Doclo, Simon

AU - Gerkmann, Timo

AU - Jensen, Søren Holdt

AU - Jensen, Jesper

PY - 2015/4/19

Y1 - 2015/4/19

N2 - In this paper we perform an extensive theoretical and experimental comparison of two recently proposed multi-channel speech dereverberation algorithms. Both of them are based on the multi-channel Wiener filter but they use different estimators of the speech and reverberation power spectral densities (PSDs). We first derive closed-form expressions for the mean square error (MSE) of both PSD estimators and then show that one estimator – previously used for speech dereverberation by the authors – always yields a better MSE. Only in the case of a two microphone array or for special spatial distributions of the interference both estimators yield the same MSE. The theoretically derived MSE values are in good agreement with numerical simulation results and with instrumental speech quality measures in a realistic speech dereverberation task for binaural hearing aids.

AB - In this paper we perform an extensive theoretical and experimental comparison of two recently proposed multi-channel speech dereverberation algorithms. Both of them are based on the multi-channel Wiener filter but they use different estimators of the speech and reverberation power spectral densities (PSDs). We first derive closed-form expressions for the mean square error (MSE) of both PSD estimators and then show that one estimator – previously used for speech dereverberation by the authors – always yields a better MSE. Only in the case of a two microphone array or for special spatial distributions of the interference both estimators yield the same MSE. The theoretically derived MSE values are in good agreement with numerical simulation results and with instrumental speech quality measures in a realistic speech dereverberation task for binaural hearing aids.

U2 - 10.1109/ICASSP.2015.7177938

DO - 10.1109/ICASSP.2015.7177938

M3 - Article in proceeding

SP - 91

EP - 95

BT - IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015

PB - IEEE Signal Processing Society

ER -

Kuklasinski A, Doclo S, Gerkmann T, Jensen SH, Jensen J. Multi-channel PSD Estimators for Speech Dereverberation: a Theoretical and Experimental Comparison. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015. IEEE Signal Processing Society. 2015. p. 91-95. (I E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings). https://doi.org/10.1109/ICASSP.2015.7177938