A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition

Speech emotion recognition is a substantial component of natural language processing (NLP). It has strict requirements for the effectiveness of feature extraction and that of the acoustic model. With that in mind, a Heterogeneous Parallel Convolution Bi-LSTM model is proposed to address the challeng...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Huiyun Zhang, Heming Huang, Henry Han
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	speech emotion recognition feature extraction heterogeneous parallel network spectral features prosodic features multi-feature fusion Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/d123ba2b76394f2eb10a3158886effba
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d123ba2b76394f2eb10a3158886effba
record_format	dspace
spelling	oai:doaj.org-article:d123ba2b76394f2eb10a3158886effba2021-11-11T15:00:09ZA Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition10.3390/app112198972076-3417https://doaj.org/article/d123ba2b76394f2eb10a3158886effba2021-10-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/21/9897https://doaj.org/toc/2076-3417Speech emotion recognition is a substantial component of natural language processing (NLP). It has strict requirements for the effectiveness of feature extraction and that of the acoustic model. With that in mind, a Heterogeneous Parallel Convolution Bi-LSTM model is proposed to address the challenges. It consists of two heterogeneous branches: the left one contains two dense layers and a Bi-LSTM layer, while the right one contains a dense layer, a convolution layer, and a Bi-LSTM layer. It can exploit the spatiotemporal information more effectively, and achieves 84.65%, 79.67%, and 56.50% unweighted average recalls on the benchmark databases EMODB, CASIA, and SAVEE, respectively. Compared with the previous research results, the proposed model achieves better performance stably.Huiyun ZhangHeming HuangHenry HanMDPI AGarticlespeech emotion recognitionfeature extractionheterogeneous parallel networkspectral featuresprosodic featuresmulti-feature fusionTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 9897, p 9897 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	speech emotion recognition feature extraction heterogeneous parallel network spectral features prosodic features multi-feature fusion Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
spellingShingle	speech emotion recognition feature extraction heterogeneous parallel network spectral features prosodic features multi-feature fusion Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999 Huiyun Zhang Heming Huang Henry Han A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition
description	Speech emotion recognition is a substantial component of natural language processing (NLP). It has strict requirements for the effectiveness of feature extraction and that of the acoustic model. With that in mind, a Heterogeneous Parallel Convolution Bi-LSTM model is proposed to address the challenges. It consists of two heterogeneous branches: the left one contains two dense layers and a Bi-LSTM layer, while the right one contains a dense layer, a convolution layer, and a Bi-LSTM layer. It can exploit the spatiotemporal information more effectively, and achieves 84.65%, 79.67%, and 56.50% unweighted average recalls on the benchmark databases EMODB, CASIA, and SAVEE, respectively. Compared with the previous research results, the proposed model achieves better performance stably.
format	article
author	Huiyun Zhang Heming Huang Henry Han
author_facet	Huiyun Zhang Heming Huang Henry Han
author_sort	Huiyun Zhang
title	A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition
title_short	A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition
title_full	A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition
title_fullStr	A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition
title_full_unstemmed	A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition
title_sort	novel heterogeneous parallel convolution bi-lstm for speech emotion recognition
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/d123ba2b76394f2eb10a3158886effba
work_keys_str_mv	AT huiyunzhang anovelheterogeneousparallelconvolutionbilstmforspeechemotionrecognition AT heminghuang anovelheterogeneousparallelconvolutionbilstmforspeechemotionrecognition AT henryhan anovelheterogeneousparallelconvolutionbilstmforspeechemotionrecognition AT huiyunzhang novelheterogeneousparallelconvolutionbilstmforspeechemotionrecognition AT heminghuang novelheterogeneousparallelconvolutionbilstmforspeechemotionrecognition AT henryhan novelheterogeneousparallelconvolutionbilstmforspeechemotionrecognition
_version_	1718437929948282880

A Novel Heterogeneous Parallel Convolution Bi-LSTM for Speech Emotion Recognition

Ejemplares similares