Multichannel parametric speech enhancement

S. Srinivasan, R. Aichner, W.B. Kleijn, W. Kellermann

Research output: Contribution to journalArticleAcademicpeer-review

12 Citations (Scopus)
79 Downloads (Pure)

Abstract

We present a parametric model-based multichannel approach for speech enhancement. By employing an autoregressive model for the speech signal and using a trained codebook of speech linear predictive coefficients, minimum mean square error estimation of the speech signal is performed. By explicitly accounting for steering errors in the signal model, robust estimates are obtained. Experiments show that the proposed method results in significant performance gains.
Original languageEnglish
Pages (from-to)304-307
Number of pages4
JournalIEEE Signal Processing Letters
Volume13
Issue number5
DOIs
Publication statusPublished - 2006

Fingerprint

Speech Enhancement
Speech enhancement
Speech Signal
Robust Estimate
Codebook
Minimum Mean Square Error
Error Estimation
Autoregressive Model
Parametric Model
Model-based
Mean square error
Error analysis
Coefficient
Experiment
Experiments
Model
Speech

Cite this

Srinivasan, S., Aichner, R., Kleijn, W. B., & Kellermann, W. (2006). Multichannel parametric speech enhancement. IEEE Signal Processing Letters, 13(5), 304-307. https://doi.org/10.1109/LSP.2005.863819
Srinivasan, S. ; Aichner, R. ; Kleijn, W.B. ; Kellermann, W. / Multichannel parametric speech enhancement. In: IEEE Signal Processing Letters. 2006 ; Vol. 13, No. 5. pp. 304-307.
@article{9615d57aaeda489d9184ffcb4d8d7e15,
title = "Multichannel parametric speech enhancement",
abstract = "We present a parametric model-based multichannel approach for speech enhancement. By employing an autoregressive model for the speech signal and using a trained codebook of speech linear predictive coefficients, minimum mean square error estimation of the speech signal is performed. By explicitly accounting for steering errors in the signal model, robust estimates are obtained. Experiments show that the proposed method results in significant performance gains.",
author = "S. Srinivasan and R. Aichner and W.B. Kleijn and W. Kellermann",
year = "2006",
doi = "10.1109/LSP.2005.863819",
language = "English",
volume = "13",
pages = "304--307",
journal = "IEEE Signal Processing Letters",
issn = "1070-9908",
publisher = "Institute of Electrical and Electronics Engineers",
number = "5",

}

Srinivasan, S, Aichner, R, Kleijn, WB & Kellermann, W 2006, 'Multichannel parametric speech enhancement', IEEE Signal Processing Letters, vol. 13, no. 5, pp. 304-307. https://doi.org/10.1109/LSP.2005.863819

Multichannel parametric speech enhancement. / Srinivasan, S.; Aichner, R.; Kleijn, W.B.; Kellermann, W.

In: IEEE Signal Processing Letters, Vol. 13, No. 5, 2006, p. 304-307.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Multichannel parametric speech enhancement

AU - Srinivasan, S.

AU - Aichner, R.

AU - Kleijn, W.B.

AU - Kellermann, W.

PY - 2006

Y1 - 2006

N2 - We present a parametric model-based multichannel approach for speech enhancement. By employing an autoregressive model for the speech signal and using a trained codebook of speech linear predictive coefficients, minimum mean square error estimation of the speech signal is performed. By explicitly accounting for steering errors in the signal model, robust estimates are obtained. Experiments show that the proposed method results in significant performance gains.

AB - We present a parametric model-based multichannel approach for speech enhancement. By employing an autoregressive model for the speech signal and using a trained codebook of speech linear predictive coefficients, minimum mean square error estimation of the speech signal is performed. By explicitly accounting for steering errors in the signal model, robust estimates are obtained. Experiments show that the proposed method results in significant performance gains.

U2 - 10.1109/LSP.2005.863819

DO - 10.1109/LSP.2005.863819

M3 - Article

VL - 13

SP - 304

EP - 307

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

SN - 1070-9908

IS - 5

ER -

Srinivasan S, Aichner R, Kleijn WB, Kellermann W. Multichannel parametric speech enhancement. IEEE Signal Processing Letters. 2006;13(5):304-307. https://doi.org/10.1109/LSP.2005.863819