Multichannel Two-Stage Beamforming with Unconstrained Beamformer and Distortion Reduction

Masahito TOGAMI; Yohei KAWAGUCHI; Yasunari OBUCHI

doi:10.1587/transfun.E96.A.749

Multichannel Two-Stage Beamforming with Unconstrained Beamformer and Distortion Reduction

Masahito TOGAMI, Yohei KAWAGUCHI, Yasunari OBUCHI

Full Text Views

0

Cite this

Summary :

This paper proposes a novel multichannel speech enhancement technique for reverberant rooms that is effective when noise sources are spatially stationary, such as a projector fan noise, an air-conditioner noise, and unwanted speech sources at the back of microphones. Speech enhancement performance of the conventional multichannel Wiener filter (MWF) degrades when the Signal-to-Noise Ratio (SNR) of the current microphone input signal changes from the noise-only period. Furthermore, the MWF structure is computationally inefficient, because the MWF updates the whole spatial beamformer periodically to track switching of the speakers (e.g. turn-taking). In contrast to the MWF, the proposed method reduces noise independently of the SNR. The proposed method has a novel two-stage structure, which reduces noise and distortion of the desired source signal in a cascade manner by using two different beamformers. The first beamformer focuses on noise reduction without any constraint on the desired source, which is insensitive to SNR variation. However, the output signal after the first beamformer is distorted. The second beamformer focuses on distortion reduction of the desired source signal. Theoretically, complete elimination of distortion is assured. Additionally, the proposed method has a computationally efficient structure optimized for spatially stationary noise reduction problems. The first beamformer is updated only when the speech enhancement system is initialized. Only the second beamformer is updated periodically to track switching of the active speaker. The experimental results indicate that the proposed method can reduce spatially stationary noise source signals effectively with less distortion of the desired source signal even in a reverberant conference room.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E96-A No.4 pp.749-761

Publication Date: 2013/04/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E96.A.749

Type of Manuscript: PAPER

Category: Engineering Acoustics

Cite this

Copy

Masahito TOGAMI, Yohei KAWAGUCHI, Yasunari OBUCHI, "Multichannel Two-Stage Beamforming with Unconstrained Beamformer and Distortion Reduction" in IEICE TRANSACTIONS on Fundamentals, vol. E96-A, no. 4, pp. 749-761, April 2013, doi: 10.1587/transfun.E96.A.749.
Abstract: This paper proposes a novel multichannel speech enhancement technique for reverberant rooms that is effective when noise sources are spatially stationary, such as a projector fan noise, an air-conditioner noise, and unwanted speech sources at the back of microphones. Speech enhancement performance of the conventional multichannel Wiener filter (MWF) degrades when the Signal-to-Noise Ratio (SNR) of the current microphone input signal changes from the noise-only period. Furthermore, the MWF structure is computationally inefficient, because the MWF updates the whole spatial beamformer periodically to track switching of the speakers (e.g. turn-taking). In contrast to the MWF, the proposed method reduces noise independently of the SNR. The proposed method has a novel two-stage structure, which reduces noise and distortion of the desired source signal in a cascade manner by using two different beamformers. The first beamformer focuses on noise reduction without any constraint on the desired source, which is insensitive to SNR variation. However, the output signal after the first beamformer is distorted. The second beamformer focuses on distortion reduction of the desired source signal. Theoretically, complete elimination of distortion is assured. Additionally, the proposed method has a computationally efficient structure optimized for spatially stationary noise reduction problems. The first beamformer is updated only when the speech enhancement system is initialized. Only the second beamformer is updated periodically to track switching of the active speaker. The experimental results indicate that the proposed method can reduce spatially stationary noise source signals effectively with less distortion of the desired source signal even in a reverberant conference room.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E96.A.749/_p

Copy

@ARTICLE{e96-a_4_749,
author={Masahito TOGAMI, Yohei KAWAGUCHI, Yasunari OBUCHI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Multichannel Two-Stage Beamforming with Unconstrained Beamformer and Distortion Reduction},
year={2013},
volume={E96-A},
number={4},
pages={749-761},
abstract={This paper proposes a novel multichannel speech enhancement technique for reverberant rooms that is effective when noise sources are spatially stationary, such as a projector fan noise, an air-conditioner noise, and unwanted speech sources at the back of microphones. Speech enhancement performance of the conventional multichannel Wiener filter (MWF) degrades when the Signal-to-Noise Ratio (SNR) of the current microphone input signal changes from the noise-only period. Furthermore, the MWF structure is computationally inefficient, because the MWF updates the whole spatial beamformer periodically to track switching of the speakers (e.g. turn-taking). In contrast to the MWF, the proposed method reduces noise independently of the SNR. The proposed method has a novel two-stage structure, which reduces noise and distortion of the desired source signal in a cascade manner by using two different beamformers. The first beamformer focuses on noise reduction without any constraint on the desired source, which is insensitive to SNR variation. However, the output signal after the first beamformer is distorted. The second beamformer focuses on distortion reduction of the desired source signal. Theoretically, complete elimination of distortion is assured. Additionally, the proposed method has a computationally efficient structure optimized for spatially stationary noise reduction problems. The first beamformer is updated only when the speech enhancement system is initialized. Only the second beamformer is updated periodically to track switching of the active speaker. The experimental results indicate that the proposed method can reduce spatially stationary noise source signals effectively with less distortion of the desired source signal even in a reverberant conference room.},
keywords={},
doi={10.1587/transfun.E96.A.749},
ISSN={1745-1337},
month={April},}

Copy

TY - JOUR
TI - Multichannel Two-Stage Beamforming with Unconstrained Beamformer and Distortion Reduction
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 749
EP - 761
AU - Masahito TOGAMI
AU - Yohei KAWAGUCHI
AU - Yasunari OBUCHI
PY - 2013
DO - 10.1587/transfun.E96.A.749
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E96-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2013
AB - This paper proposes a novel multichannel speech enhancement technique for reverberant rooms that is effective when noise sources are spatially stationary, such as a projector fan noise, an air-conditioner noise, and unwanted speech sources at the back of microphones. Speech enhancement performance of the conventional multichannel Wiener filter (MWF) degrades when the Signal-to-Noise Ratio (SNR) of the current microphone input signal changes from the noise-only period. Furthermore, the MWF structure is computationally inefficient, because the MWF updates the whole spatial beamformer periodically to track switching of the speakers (e.g. turn-taking). In contrast to the MWF, the proposed method reduces noise independently of the SNR. The proposed method has a novel two-stage structure, which reduces noise and distortion of the desired source signal in a cascade manner by using two different beamformers. The first beamformer focuses on noise reduction without any constraint on the desired source, which is insensitive to SNR variation. However, the output signal after the first beamformer is distorted. The second beamformer focuses on distortion reduction of the desired source signal. Theoretically, complete elimination of distortion is assured. Additionally, the proposed method has a computationally efficient structure optimized for spatially stationary noise reduction problems. The first beamformer is updated only when the speech enhancement system is initialized. Only the second beamformer is updated periodically to track switching of the active speaker. The experimental results indicate that the proposed method can reduce spatially stationary noise source signals effectively with less distortion of the desired source signal even in a reverberant conference room.
ER -