New Generalized Sidelobe Canceller with Denoising Auto-Encoder for Improved Speech Enhancement

Minkyu SHIN; Seongkyu MUN; David K. HAN; Hanseok KO

doi:10.1587/transfun.E100.A.3038

New Generalized Sidelobe Canceller with Denoising Auto-Encoder for Improved Speech Enhancement

Minkyu SHIN, Seongkyu MUN, David K. HAN, Hanseok KO

Full Text Views

0

Share
Cite this

Summary :

In this paper, a multichannel speech enhancement system which adopts a denoising auto-encoder as part of the beamformer is proposed. The proposed structure of the generalized sidelobe canceller generates enhanced multi-channel signals, instead of merely one channel, to which the following denoising auto-encoder can be applied. Because the beamformer exploits spatial information and compensates for differences in the transfer functions of each channel, the proposed system is expected to resolve the difficulty of modelling relative transfer functions consisting of complex numbers which are hard to model with a denoising auto-encoder. As a result, the modelling capability of the denoising auto-encoder can concentrate on removing the artefacts caused by the beamformer. Unlike conventional beamformers, which combine these artefacts into one channel, they remain separated for each channel in the proposed method. As a result, the denoising auto-encoder can remove the artefacts by referring to other channels. Experimental results prove that the proposed structure is effective for the six-channel data in CHiME, as indicated by improvements in terms of speech enhancement and word error rate in automatic speech recognition.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E100-A No.12 pp.3038-3040

Publication Date: 2017/12/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E100.A.3038

Type of Manuscript: LETTER

Category: Speech and Hearing

Authors

Minkyu SHIN
  Korea University
Seongkyu MUN
  Korea University
David K. HAN
  the Office of Naval Research
Hanseok KO
  Korea University

Keyword

speech enhancement, denoising auto-encoder, acoustic beamforming, generalized sidelobe canceller

Cite this

Copy

Minkyu SHIN, Seongkyu MUN, David K. HAN, Hanseok KO, "New Generalized Sidelobe Canceller with Denoising Auto-Encoder for Improved Speech Enhancement" in IEICE TRANSACTIONS on Fundamentals, vol. E100-A, no. 12, pp. 3038-3040, December 2017, doi: 10.1587/transfun.E100.A.3038.
Abstract: In this paper, a multichannel speech enhancement system which adopts a denoising auto-encoder as part of the beamformer is proposed. The proposed structure of the generalized sidelobe canceller generates enhanced multi-channel signals, instead of merely one channel, to which the following denoising auto-encoder can be applied. Because the beamformer exploits spatial information and compensates for differences in the transfer functions of each channel, the proposed system is expected to resolve the difficulty of modelling relative transfer functions consisting of complex numbers which are hard to model with a denoising auto-encoder. As a result, the modelling capability of the denoising auto-encoder can concentrate on removing the artefacts caused by the beamformer. Unlike conventional beamformers, which combine these artefacts into one channel, they remain separated for each channel in the proposed method. As a result, the denoising auto-encoder can remove the artefacts by referring to other channels. Experimental results prove that the proposed structure is effective for the six-channel data in CHiME, as indicated by improvements in terms of speech enhancement and word error rate in automatic speech recognition.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/transfun.E100.A.3038/_p

Copy

@ARTICLE{e100-a_12_3038,
author={Minkyu SHIN, Seongkyu MUN, David K. HAN, Hanseok KO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={New Generalized Sidelobe Canceller with Denoising Auto-Encoder for Improved Speech Enhancement},
year={2017},
volume={E100-A},
number={12},
pages={3038-3040},
abstract={In this paper, a multichannel speech enhancement system which adopts a denoising auto-encoder as part of the beamformer is proposed. The proposed structure of the generalized sidelobe canceller generates enhanced multi-channel signals, instead of merely one channel, to which the following denoising auto-encoder can be applied. Because the beamformer exploits spatial information and compensates for differences in the transfer functions of each channel, the proposed system is expected to resolve the difficulty of modelling relative transfer functions consisting of complex numbers which are hard to model with a denoising auto-encoder. As a result, the modelling capability of the denoising auto-encoder can concentrate on removing the artefacts caused by the beamformer. Unlike conventional beamformers, which combine these artefacts into one channel, they remain separated for each channel in the proposed method. As a result, the denoising auto-encoder can remove the artefacts by referring to other channels. Experimental results prove that the proposed structure is effective for the six-channel data in CHiME, as indicated by improvements in terms of speech enhancement and word error rate in automatic speech recognition.},
keywords={},
doi={10.1587/transfun.E100.A.3038},
ISSN={1745-1337},
month={December},}

Copy

TY - JOUR
TI - New Generalized Sidelobe Canceller with Denoising Auto-Encoder for Improved Speech Enhancement
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 3038
EP - 3040
AU - Minkyu SHIN
AU - Seongkyu MUN
AU - David K. HAN
AU - Hanseok KO
PY - 2017
DO - 10.1587/transfun.E100.A.3038
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E100-A
IS - 12
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - December 2017
AB - In this paper, a multichannel speech enhancement system which adopts a denoising auto-encoder as part of the beamformer is proposed. The proposed structure of the generalized sidelobe canceller generates enhanced multi-channel signals, instead of merely one channel, to which the following denoising auto-encoder can be applied. Because the beamformer exploits spatial information and compensates for differences in the transfer functions of each channel, the proposed system is expected to resolve the difficulty of modelling relative transfer functions consisting of complex numbers which are hard to model with a denoising auto-encoder. As a result, the modelling capability of the denoising auto-encoder can concentrate on removing the artefacts caused by the beamformer. Unlike conventional beamformers, which combine these artefacts into one channel, they remain separated for each channel in the proposed method. As a result, the denoising auto-encoder can remove the artefacts by referring to other channels. Experimental results prove that the proposed structure is effective for the six-channel data in CHiME, as indicated by improvements in terms of speech enhancement and word error rate in automatic speech recognition.
ER -