[1]李田港,叶 硕,叶光明,等.基于集成学习的语音情感识别算法研究[J].计算机技术与发展,2020,30(06):82-86.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 016]
　LI Tian-gang,YE Shuo,YE Guang-ming,et al.Research on Speech Emotion Recognition Algorithm Based on Ensemble Learning[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2020,30(06):82-86.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 016]
点击复制

基于集成学习的语音情感识别算法研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年06期
页码:
82-86
栏目:
智能、算法、系统工程
出版日期:
2020-06-10

文章信息/Info

Title:
Research on Speech Emotion Recognition Algorithm Based on Ensemble Learning
文章编号:
1673-629X(2020)06-0082-05
作者:
李田港1 ; 叶 硕1 ; 叶光明2 ; 褚 钰1
1. 武汉邮电科学研究院,湖北 武汉 430000; 2. 武汉烽火众智数字技术有限责任公司,湖北 武汉 430000
Author(s):
LI Tian-gang1 ; YE Shuo1 ; YE Guang-ming2 ; CHU Yu1
1. Wuhan Research Institute of Posts and Telecommunications,Wuhan 430000,China; 2. Wuhan Fiberhome Wisdom Digital Technology Co. ,Ltd. ,Wuhan 430000,China
关键词:
语音识别; 情感识别; SVM; W-KNN; BPNN; 集成学习
Keywords:
speech recognition; emotion recognition; SVM; W-KNN; BPNN; ensemble learning
分类号:
TP301. 6
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 06. 016
摘要:
语音情感识别是语音识别的热门方向,心理学将情感识别分为离散型和连续型,离散型情感识别常用的声学特征为韵律学特征、基于谱的相关特征、音质特征,识别方法通常有 KNN、SVM、HMM 等。 提出一种基于距离加权的改进 KNN 算法,引入类平均距离作为加权依据,并设计一种基于集成学习的加权投票算法,将改进 KNN、SVM、BPNN 分类方法进行集成,提高语音情感识别率。 实验表明,改进后的 KNN 算法相比传统 KNN,识别率在不同语种的语料库上均有提升,最大提升为 9. 6% ,且表现结果稳定,准确率与 SVM、BPNN 大致相当,可用于集成学习;对比单一识别算法,所设计的集成学习算法具有较高可靠性,在生气、高兴、悲伤、惊慌及中性情感上均达到较好的识别效果,实现了离散型语音情感的识别。
Abstract:
Speech emotion recognition is a popular direction of speech recognition. Psychology divides emotional recognition into discrete and continuous.For discrete emotion recognition,the acoustic features commonly used are prosodic features, spectral related features,and acoustic quality features. The recognition methods usually include KNN,SVM, HMM, etc. We propose an improved KNN algorithm based on distance weighting,introduce the class average distance as the weighting basis,and design a weighted voting algorithm based on ensemble learning, which integrates the improved KNN, SVM and BPNN classification methods to improve the speech emotion recognition rate. Experiments show that compared with traditional KNN,the recognition rate of the improved KNN algorithm is improved on corpus of different languages,with the maximum rate of 9.6% and stable performance result. The accuracy rate is roughly the same as SVM and BPNN,which can be used for ensemble learning. Compared with the single recognition algorithm, the designed ensemble learning algorithm has higher reliability,achieves better recognition effect in anger,happiness, sadness, panic and neutral emotion, and realizes the recognition of discrete speech emotion.

相似文献/References:

[1]宋鑫坤 陈万米 朱明 桂春胜 程硕远 陈海波.基于正则表达式的语音识别控制策略研究[J].计算机技术与发展,2010,(02):106.
　SONG Xin-kun,CHEN Wan-mi,ZHU Ming,et al.Study on Speech Recognition Control Strategy Based on Regular Expression[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2010,(06):106.
[2]石现峰 张学智 张峰.基于HTK的语音识别系统设计[J].计算机技术与发展,2006,(10):37.
　SHI Xian-feng,ZHANG Xue-zhi,ZHANG Feng.Design of Speech Recognition System Based on HTK[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2006,(06):37.
[3]朱宇 宋艳.嵌入式语音识别系统特征参数提取研究[J].计算机技术与发展,2011,(07):246.
　ZHU Yu,SONG Yan.Research of Characteristic Parameters Extraction Based on Embedded Speech Recognition System[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2011,(06):246.
[4]林鸣霄.基于SpeechSDK的语音识别技术在三维仿真中的应用[J].计算机技术与发展,2011,(11):160.
　LIN Ming-xiao.Application of Speech Recognition Technology in 3D Simulation Based on Speech SDK[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2011,(06):160.
[5]韩志艳,伦淑娴,王健.基于遗传小波神经网络的语音情感识别[J].计算机技术与发展,2013,(01):75.
　HAN Zhi-yan,LUN Shu-xian,WANG Jian.Speech Emotion Recognition Based on Genetic Wavelet Neural Network[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2013,(06):75.
[6]李克粉,王直.改进的小波阈值去噪在语音识别中的应用[J].计算机技术与发展,2013,(05):231.
　LI Ke-fen,WANG Zhi.Application of Improved Wavelet Threshold Denoising in Speech Recognition[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2013,(06):231.
[7]王海洋,郭星. 基于语音识别的智慧旅游系统研究[J].计算机技术与发展,2015,25(05):143.
　WANG Hai-yang,GUO Xing. Study on Smart Tourism System Based on Voice Recognition[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2015,25(06):143.
[8]孙科学[] [],洪櫆[],章康宁[],等. 一种联合检测门禁系统的设计与实现[J].计算机技术与发展,2016,26(01):155.
　SUN Ke-xue[][],HONG Kui[],ZHANG Kang-ning[],et al. Design and Implementation of Joint Detection Access Control System[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2016,26(06):155.
[9]韩志艳,王健. 多模式情感识别特征参数融合算法研究[J].计算机技术与发展,2016,26(05):27.
　HAN Zhi-yan,WANG Jian. Research on Feature Fusion Algorithm for Multimodal Emotion Recognition[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2016,26(06):27.
[10]李姗,徐珑婷. 基于语谱图提取瓶颈特征的情感识别算法研究[J].计算机技术与发展,2017,27(05):82.
　LI Shan,XU Long-ting. Research on Emotion Recognition Algorithm Based on Spectrogram Feature Extraction of Bottleneck Feature[J].COMPUTER TECHNOLOGY AND DEVELOPMENT,2017,27(06):82.

更新日期/Last Update: 2020-06-10