On Optimal Frequency-Domain Multichannel Linear Filtering for Noise Reduction

Souden, Mehrez; Benesty, Jacob; Affes, Sofiène

doi:10.1109/tasl.2009.2025790

Cited by 253 publications

(183 citation statements)

References 34 publications

Supporting

Mentioning

183

Contrasting

Order By: Relevance

“…From these expressions we also see that the noise-reduction factor of the MVDR and LCMV are similar when is much larger than . Finally, the narrowband noise reduction for the PMWF is given by [7] (68) with (69) Using the matrix inversion lemma it can be shown that for the considered scenario (70) Analyzing (68) we find that for a given situation and hence and , the noise reduction of the PMWF increases monotonically with .…”

Section: B Noise Reductionmentioning

confidence: 96%

“…In earlier work the speech distortion was controlled by augmenting the classical meansquare error (MSE) cost function [6] or by imposing a linear inequality constraint [7], [16], [17]. It should be noted that the inequality constraint on the PSD of the desired signal at the beamformer's output (as used in [7], [17]) is binding. Consequently, the same solution is obtained if the inequality constraint is replaced by an equality constraint.…”

Section: Speech-distortion and Interference-rejection Constraintmentioning

confidence: 99%

“…Finally, we show the relation between the SDIRC beamformer and the the PMWF that is given by [7] (35) To obtain results with the SDIRC beamformer equal to those obtained by the parameterized MWF we require in addition to and the RTFs and .…”

Section: Pmwfmentioning

confidence: 99%

“…The multichannel Wiener filter (MWF), on the other hand, reduces the interference-plus-noise power without exactly preserving the desired signal. In order to control the amount of distortion, the parameterized multichannel Wiener filter (also know as speech-distortion weighted multichannel Wiener filter) has been proposed [6], [7].…”

mentioning

confidence: 99%

“…In [13], Bitzer et al investigated the theoretical performance limits of the GSC in the case of a spatially diffuse noise. The tradeoff between noise reduction and speech distortion in the parameterized multichannel Wiener filtering was established in [7]. Another notable effort to understand the functioning of the TF ratios-based GSC beamformer was published by Gannot and Cohen in [2].…”

mentioning

confidence: 99%

See 4 more Smart Citations

A Speech Distortion and Interference Rejection Constraint Beamformer

Habets

Benesty

Naylor

2012

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-Signals captured by a set of microphones in a speech communication system are mixtures of desired and undesired signals and ambient noise. Existing beamformers can be divided into those that preserve or distort the desired signal. Beamformers that preserve the desired signal are, for example, the linearly constrained minimum variance (LCMV) beamformer that is supposed, ideally, to reject the undesired signal and reduce the ambient noise power, and the minimum variance distortionless response (MVDR) beamformer that reduces the interference-plus-noise power. The multichannel Wiener filter, on the other hand, reduces the interference-plus-noise power without preserving the desired signal. In this paper, a speech distortion and interference rejection constraint (SDIRC) beamformer is derived that minimizes the ambient noise power subject to specific constraints that allow a tradeoff between speech distortion and interference-plus-noise reduction on the one hand, and undesired signal and ambient noise reductions on the other hand. Closed-form expressions for the performance measures of the SDIRC beamformer are derived and the relations to the aforementioned beamformers are derived. The performance evaluation demonstrates the tradeoffs that can be made using the SDIRC beamformer.Index Terms-Beamforming, frequency domain, linearly constrained minimum variance (LCMV), minimum variance distortionless response (MVDR), noise reduction, parameterized multichannel Wiener filter, speech enhancement.

show abstract

Section: B Noise Reductionmentioning

confidence: 96%

Section: Speech-distortion and Interference-rejection Constraintmentioning

confidence: 99%

Section: Pmwfmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

A Speech Distortion and Interference Rejection Constraint Beamformer

Habets

Benesty

Naylor

2012

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

DNNBased Speech Quality Enhancement and Multi‐speaker Separation for Automatic Speech Recognition System

Ramya¹,

Sakthi²

2022

Machine Learning Algorithms for Signal and Image Processing

View full text Add to dashboard Cite

Distributed multichannel Wiener filtering for speech enhancement in acoustic sensor networks

Chang

Chen

Yin

2022

Adaptive Control & Signal

View full text Add to dashboard Cite

Summary The performance of speech processing algorithms in wireless acoustic sensor networks (WASNs) is severely degraded by diffuse noise. To address this problem, a distributed speech enhancement algorithm for WASNs is proposed in this article. The parameterized multichannel Wiener filter is first improved by exploiting all cross‐correlation information among nodes. Next, using the Hadamard element‐wise product of label matrix and cross‐correlation matrix, the distributed parameterized multichannel Wiener filter based on the in‐network summation is presented. Moreover, by making the tradeoff between communication cost and enhancement performance, a simplified distributed algorithm is proposed. Finally, the noise reduction performance of the proposed method is theoretically analyzed. The proposed method achieves good speech enhancement performance and requires only in‐network summation in WASNs. Experimental results reveal the validity of the proposed method.

show abstract

On Optimal Frequency-Domain Multichannel Linear Filtering for Noise Reduction

Cited by 253 publications

References 34 publications

A Speech Distortion and Interference Rejection Constraint Beamformer

A Speech Distortion and Interference Rejection Constraint Beamformer

DNNBased Speech Quality Enhancement and Multi‐speaker Separation for Automatic Speech Recognition System

Distributed multichannel Wiener filtering for speech enhancement in acoustic sensor networks

Contact Info

Product

Resources

About