Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism

Speech enhancement in a vehicle environment remains a challenging task for the complex noise. The paper presents a feature extraction method that we use interchannel attention mechanism frame by frame for learning spatial features directly from the multichannel speech waveforms. The spatial features...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xueli Shen, Zhenxing Liang, Shiyin Li, Yanji Jiang
Formato:	article
Lenguaje:	EN
Publicado:	Hindawi-Wiley 2021
Materias:	Transportation engineering TA1001-1280 Transportation and communications HE1-9990
Acceso en línea:	https://doaj.org/article/e7c29eef5ac44c39ac0fc2dfc3aeb7fd
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e7c29eef5ac44c39ac0fc2dfc3aeb7fd
record_format	dspace
spelling	oai:doaj.org-article:e7c29eef5ac44c39ac0fc2dfc3aeb7fd2021-11-29T00:57:05ZMultichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism2042-319510.1155/2021/9453911https://doaj.org/article/e7c29eef5ac44c39ac0fc2dfc3aeb7fd2021-01-01T00:00:00Zhttp://dx.doi.org/10.1155/2021/9453911https://doaj.org/toc/2042-3195Speech enhancement in a vehicle environment remains a challenging task for the complex noise. The paper presents a feature extraction method that we use interchannel attention mechanism frame by frame for learning spatial features directly from the multichannel speech waveforms. The spatial features of the individual signals learned through the proposed method are provided as an input so that the two-stage BiLSTM network is trained to perform adaptive spatial filtering as time-domain filters spanning signal channels. The two-stage BiLSTM network is capable of local and global features extracting and reaches competitive results. Using scenarios and data based on car cockpit simulations, in contrast to other methods that extract the feature from multichannel data, the results show the proposed method has a significant performance in terms of all SDR, SI-SNR, PESQ, and STOI.Xueli ShenZhenxing LiangShiyin LiYanji JiangHindawi-WileyarticleTransportation engineeringTA1001-1280Transportation and communicationsHE1-9990ENJournal of Advanced Transportation, Vol 2021 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Transportation engineering TA1001-1280 Transportation and communications HE1-9990
spellingShingle	Transportation engineering TA1001-1280 Transportation and communications HE1-9990 Xueli Shen Zhenxing Liang Shiyin Li Yanji Jiang Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism
description	Speech enhancement in a vehicle environment remains a challenging task for the complex noise. The paper presents a feature extraction method that we use interchannel attention mechanism frame by frame for learning spatial features directly from the multichannel speech waveforms. The spatial features of the individual signals learned through the proposed method are provided as an input so that the two-stage BiLSTM network is trained to perform adaptive spatial filtering as time-domain filters spanning signal channels. The two-stage BiLSTM network is capable of local and global features extracting and reaches competitive results. Using scenarios and data based on car cockpit simulations, in contrast to other methods that extract the feature from multichannel data, the results show the proposed method has a significant performance in terms of all SDR, SI-SNR, PESQ, and STOI.
format	article
author	Xueli Shen Zhenxing Liang Shiyin Li Yanji Jiang
author_facet	Xueli Shen Zhenxing Liang Shiyin Li Yanji Jiang
author_sort	Xueli Shen
title	Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism
title_short	Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism
title_full	Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism
title_fullStr	Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism
title_full_unstemmed	Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism
title_sort	multichannel speech enhancement in vehicle environment based on interchannel attention mechanism
publisher	Hindawi-Wiley
publishDate	2021
url	https://doaj.org/article/e7c29eef5ac44c39ac0fc2dfc3aeb7fd
work_keys_str_mv	AT xuelishen multichannelspeechenhancementinvehicleenvironmentbasedoninterchannelattentionmechanism AT zhenxingliang multichannelspeechenhancementinvehicleenvironmentbasedoninterchannelattentionmechanism AT shiyinli multichannelspeechenhancementinvehicleenvironmentbasedoninterchannelattentionmechanism AT yanjijiang multichannelspeechenhancementinvehicleenvironmentbasedoninterchannelattentionmechanism
_version_	1718407681563164672

Multichannel Speech Enhancement in Vehicle Environment Based on Interchannel Attention Mechanism

Ejemplares similares