[1]俞颖,黄风华,刘永芬.基于特征降维及参数优化的语音情感识别[J].延边大学学报(自然科学版),2020,46(01):49-54.
　YU Ying,HUANG Fenghua,LIU Yongfen.Speech emotion recognition based on feature dimension reduction and parameter optimization[J].Journal of Yanbian University,2020,46(01):49-54.
点击复制

基于特征降维及参数优化的语音情感识别

《延边大学学报(自然科学版)》[ISSN:1004-4353/CN:22-1191/N] 卷: 第46卷 期数: 2020年01期 页码: 49-54 栏目: 基础科学研究 出版日期: 2020-05-26
Title:
Speech emotion recognition based on feature dimension reduction and parameter optimization
文章编号:
1004-4353(2020)01-0049-06
作者:
俞颖1; 2;  黄风华1; 2;  刘永芬3
( 1.阳光学院 空间数据挖掘与应用福建省高校工程研究中心; 2.阳光学院 人工智能学院; 3.福建农林大学 金山学院: 福建 福州 350001 )
Author(s):
YU Ying1; 2;  HUANG Fenghua1; 2;  LIU Yongfen3
( 1.Spatial Data Mining and Application Research Center of Fujian Province, Yango University; 2.Artificial Intelligence College, Yango University; 3.Jinshan College, Fujian Agriculture and Forestry University: Fuzhou 350001, China )
关键词:
快速主成份分析法;  遗传算法;  BP神经网络;  语音情感识别
Keywords:
Fast_PAC;  genetic algorithm;  BP neural network;  speech emotion recognition
分类号:
TP301.6
文献标志码:
A
摘要:
针对传统BP神经网络在语音情感识别过程中存在的计算量偏大和容易陷入局部最优解的问题,提出了一种基于特征降维及参数优化的情感识别改进方法.首先提取情感语料库的高维度联合特征,利用快速主成份分析法(Fast_PAC)进行特征降维以达到降低问题复杂性的目的; 然后引入遗传算法对BP神经网络进行参数优化以避免限入局部最优问题; 最后构建语音情感识别分类器,并利用CASIA汉语语料库及柏林德语语料库进行情感识别验证.验证结果表明,与传统的支持向量机(SVM)方法、传统的主成份分析法(PCA算法)结合SVM模型识别方
Abstract:
The traditional BP neural network has been existing some burning questions in the process of speech emotion recognition, especially the high computational and local optimum trending. Against these shortcomings, we present a novel method of emotion recognition based on feature dimension reduction and parameter optimization. The recognition method is divided into three stages. In the first stage, it extracts the high-dimensional joint features of the speech emotion database. This is, in fact, aimingto reduce the complexity of the problem which is carried out by the fast principal component analysis(Fast_PAC)method. In the second stage, genetic algorithm is used to optimize the parameters of BP neural network to avoid the local optimum problem. Finally, we construct a speech emotion recognition classifier, and take the experiments on the CASIA Chinese corpus and Berlin German corpus for emotion recognition verification. The experiments show that the proposed method can effectively reduce the feature dimension of speech emotion comparing with other competitive methods, such as the traditional support vector machine(SVM)method and the traditional PCA combined with SVM model recognition method. Furthermore, it demonstrates the advantages of less computation and higher recognition accuracy.

参考文献/References:

[1] 王富,孙林慧,苏敏,等.基于参数寻优决策树SVM的语音情感识别[J].计算机技术与发展,2018,28(7):63-65.
[2] ZHU J C, LIU Z L. Analysis of hybrid feature research based on extraction LPCC and MFCC[C]//Tenth International Conference on Computational Intelligence and Security. Kunming: IEEE, 2014:732-735.
[3] 李高玲,帖云,齐林.基于随机森林分类优化的多特征语音情感识别[J].微电子学与计算机,2019,36(1):70-73.
[4] 任浩,叶亮,李月,等.基于多级SVM分类的语音情感识别算法[J].计算机应用研究,2017,34(6):1682-1684.
[5] 蒋海华,胡斌.基于PCA和SVM的普通话语音情感识别[J].计算机科学,2015,42(11):270-272.
[6] 陈闯,Ryad Chellali,邢尹.改进遗传算法优化BP神经网络的语音情感识别[J].计算机应用研究,2019,36(2):344-345.
[7] 徐照松,元建.基于BP神经网络的语音情感识别研究[J].软件导刊,2014,13(4):11-12.
[8] 崔星星,苏智剑.一种新呼吸音信号特征提取方法与应用[J].中国医学物理学杂志,2018,35(2):214-218.
[9] 刘晨轩,蓝贤桂.语音信号短时分析算法研究与实现[J].价值工程,2012,12:191-192.
[10] 李强,刘晓峰,贺静.基于语音特征的情感分类[J].小型微型计算机系统,2016,37(2):385-387.
[11] 周萍,沈昊,郑凯鹏.基于MFCC与GFCC混合特征参数的说话人识别[J].应用科学学报,2019,37(1):24-32.
[12] 廖周宇,王钰婷,谢晓兰,等.基于粒子群优化的支持向量机人脸识别[J].计算机工程,2017,43(12):248-250.
[13] MITTAL N, WALIA E. Face recognition using improved fast PCA algorithm[C]// Proceedings of the 2008 Congress on Image and Signal Processing. Sanya: IEEE Computer Society, 2008:554-558.
[14] 杨怡涵,柳炳祥.一种基于遗传算法优化BP神经网络的陶瓷原料分类方法[J].陶瓷学报,2018,39(3):340-342.
[15] Institute of Automation, Chinese Academy of Sciences. CASIA Mandarin emotional corpus[DB/OL]. [2019-10-12].http://www.chineseIdc.org/resource_info.php?rid=76Casis.
[16] BURKHARDT F, PAESCHKE A, ROLFES M, et al. Adaabase of German emotional speech[C]//Proc of Interspeech. Lisbon: ISCA, 2005:1517-1520.

相似文献/References:

[1]汪岚.基于智能混合算法的车辆配送路径优化[J].延边大学学报(自然科学版),2015,41(03):261.
　WANG Lan.Research on optimizing vehicle routing based on intelligent hybrid algorithm[J].Journal of Yanbian University,2015,41(01):261.
[2]程永强,汪岚.基于改进MMAS算法的堆垛机货物拣选路径优化[J].延边大学学报(自然科学版),2017,43(03):255.
　CHENG Yongqiang,WANG Lan.The optimization of order picking route of stackerbased on improved MMAS algorithm[J].Journal of Yanbian University,2017,43(01):255.
[3]安迪,廉哲满*.基于MATLAB遗传算法的离合器摩擦片优化设计[J].延边大学学报(自然科学版),2018,44(03):279.
　AN Di,LIAN Zheman*.The optimization design of clutch friction disks onMATLAB genetic algorithm[J].Journal of Yanbian University,2018,44(01):279.
[4]汪岚,吴永春,陈海洋.带软时间窗的公共自行车调度路径问题[J].延边大学学报(自然科学版),2019,45(01):64.
　WANG Lan,WU Yongchun,CHEN Haiyang.Scheduling routing problem with soft time windows for public bicycle[J].Journal of Yanbian University,2019,45(01):64.
[5]罗艳霞,杨云,车艳.基于遗传算法的陶瓷配方优化设计[J].延边大学学报(自然科学版),2019,45(02):166.
　LUO Yanxia,YANG Yun,CHE Yan.Design of ceramic formula based on genetic algorithms[J].Journal of Yanbian University,2019,45(01):166.

备注/Memo

收稿日期: 2019-12-26
作者简介: 俞颖(1984—),女,讲师,研究方向为数据挖掘、模式识别.
基金项目: 福建省教育厅中青年教师科研项目(JAT190977); 福建省自然科学基金资助项目(2019J01088)

更新日期/Last Update: 2020-05-26