Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network
Automatic speech emotion recognition is a challenging task due to the gap between acoustic features and human emotions, which rely strongly on the discriminative acoustic features extracted for a given recognition task. We propose a novel deep neural architecture to extract the informative feature r...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/19/12/2730 |
_version_ | 1811262569581641728 |
---|---|
author | Wei Jiang Zheng Wang Jesse S. Jin Xianfeng Han Chunguang Li |
author_facet | Wei Jiang Zheng Wang Jesse S. Jin Xianfeng Han Chunguang Li |
author_sort | Wei Jiang |
collection | DOAJ |
description | Automatic speech emotion recognition is a challenging task due to the gap between acoustic features and human emotions, which rely strongly on the discriminative acoustic features extracted for a given recognition task. We propose a novel deep neural architecture to extract the informative feature representations from the heterogeneous acoustic feature groups which may contain redundant and unrelated information leading to low emotion recognition performance in this work. After obtaining the informative features, a fusion network is trained to jointly learn the discriminative acoustic feature representation and a Support Vector Machine (SVM) is used as the final classifier for recognition task. Experimental results on the IEMOCAP dataset demonstrate that the proposed architecture improved the recognition performance, achieving accuracy of 64% compared to existing state-of-the-art approaches. |
first_indexed | 2024-04-12T19:27:54Z |
format | Article |
id | doaj.art-d11690a142a04eca80ee4f6f646b843a |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-12T19:27:54Z |
publishDate | 2019-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-d11690a142a04eca80ee4f6f646b843a2022-12-22T03:19:26ZengMDPI AGSensors1424-82202019-06-011912273010.3390/s19122730s19122730Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural NetworkWei Jiang0Zheng Wang1Jesse S. Jin2Xianfeng Han3Chunguang Li4College of Intelligence and Computing, Tianjin University, Tianjin 300072, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin 300072, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin 300072, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin 300072, ChinaSchool of Computer Information and Engineering, Changzhou Institute of Technology, Changzhou 213032, ChinaAutomatic speech emotion recognition is a challenging task due to the gap between acoustic features and human emotions, which rely strongly on the discriminative acoustic features extracted for a given recognition task. We propose a novel deep neural architecture to extract the informative feature representations from the heterogeneous acoustic feature groups which may contain redundant and unrelated information leading to low emotion recognition performance in this work. After obtaining the informative features, a fusion network is trained to jointly learn the discriminative acoustic feature representation and a Support Vector Machine (SVM) is used as the final classifier for recognition task. Experimental results on the IEMOCAP dataset demonstrate that the proposed architecture improved the recognition performance, achieving accuracy of 64% compared to existing state-of-the-art approaches.https://www.mdpi.com/1424-8220/19/12/2730human–computer interaction (HCI)speech emotion recognitiondeep neural architectureheterogeneous feature unificationfusion network |
spellingShingle | Wei Jiang Zheng Wang Jesse S. Jin Xianfeng Han Chunguang Li Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network Sensors human–computer interaction (HCI) speech emotion recognition deep neural architecture heterogeneous feature unification fusion network |
title | Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network |
title_full | Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network |
title_fullStr | Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network |
title_full_unstemmed | Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network |
title_short | Speech Emotion Recognition with Heterogeneous Feature Unification of Deep Neural Network |
title_sort | speech emotion recognition with heterogeneous feature unification of deep neural network |
topic | human–computer interaction (HCI) speech emotion recognition deep neural architecture heterogeneous feature unification fusion network |
url | https://www.mdpi.com/1424-8220/19/12/2730 |
work_keys_str_mv | AT weijiang speechemotionrecognitionwithheterogeneousfeatureunificationofdeepneuralnetwork AT zhengwang speechemotionrecognitionwithheterogeneousfeatureunificationofdeepneuralnetwork AT jessesjin speechemotionrecognitionwithheterogeneousfeatureunificationofdeepneuralnetwork AT xianfenghan speechemotionrecognitionwithheterogeneousfeatureunificationofdeepneuralnetwork AT chunguangli speechemotionrecognitionwithheterogeneousfeatureunificationofdeepneuralnetwork |