提出了一种新的基于情感特征提取和借助支持向量机(SVM)分类器(classifier)的情感互相关性算法,并应用于语音情感识别.SVM分类器是利用情感语音信号中互相关性的特征提取进行分类的.利用这种方法对4种情感(愤怒、高兴、悲伤和中立)语音信号进行情感分类.借助SVM分类器的情感互相关性算法,可以大幅提高情感识别率,并且在识别愤怒情感时准确率甚至可以达到95.04\%.
Abstract
This paper proposes a relatively new emotional cross-correlation algorithm based emotional feature extractor and is aided with support vector machine (SVM) classifier for speech emotional recognition. The SVM classifier employs suitable features extracted from cross-correlation of emotional speech signals. The proposed technique has been utilized for classification of four kinds of emotional (anger、happy、sad and neutral) speech signals. This emotional cross-correlation algorithm aided SVM classification system could considerable improve emotional recognition rate, and detects anger emotion efficiently with a recognition rate of 95.04\%.
关键词
语音情感识别 /
支持向量机 /
结构风险最小化 /
情感模式互相关性 /
情感支持向量
{{custom_keyword}} /
Key words
speech emotional recognition /
support vector machines (SVM) /
structural risk minimization /
emotional pattern cross-correlation /
emotional support vector
{{custom_keyword}} /
中图分类号：
TP391.41
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 何志国,何钦铭,陈奇,等. 语音识别中结合进化计算与MDI 的HMM训练方法［J］. 系统工程理论与实践, 2006, 26(7): 54-58. He Z G, He Q M, Chen Q, et al. A HMM training method based on evolutionary computation and MDI in speech recognition［J］. Systems Engineering —— Theory & Practice, 2006, 26(7): 54-58.
[2] Lin Y L, Wei G. Speech emotion recognition based on HMM and SVM［C］//Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, 2005: 4898-4901.
[3] Cristianini N, Shawe-Taylor J.支持向量机导论［M］.北京: 电子工业出版社, 2004: 78-99. Cristianini N, Shawe-Taylor J. An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods［M］. Beijing: Publishing House of Electronics Industry, 2004: 78-99.
[4] Quatieri T F. Discrete-Time Speech Signal Processing: Principles and Practice［M］. Beijing: Publishing House of Electronics Industry, 2004: 468-488.
[5] Deng N Y, Tian Y J. Support Vector Machines: Principles, Algorithm and Expand［M］. Beijing: Science Press, 2009: 131-156.
[6] Begg R K, Palaniswami M, Owen B. Support vector machines for automated gait classification［J］. IEEE Trans, Biomed, Eng, 2005, 52(5): 828-838.
[7] Foody G M, Mathur A. A relative evaluation of multiclass image classification by support vector machines［J］. IEEE Trans, Geosci, Remote Sens, 2004, 42(6): 1335-1343.
[8] Cernak M, Wellekens C. Emotional aspects of intrinsic speech variabilities in automatic speech recognition［C］// International Conference on Speech and Computer, 2006: 405-408.
[9] Lugger M, Yang B. Psychological Motivated Multi-Stage Emotion Classification Exploiting Voice Quality Features［M］// Mihelic F, Zibert J. Speech Recognition, In-Tech, 2008: Chapter22.
[10] Xu Y, Chuenwattanapranithi S. Perceiving anger and joy in speech through the size code［C］//Proceedings of the International Conference on Phonetic Sciences, 2007: 2105-2108.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
山西省国际科技合作项目(2011081047);山西省自然科学基金(2010011020-1)
{{custom_fund}}
PDF(1156 KB)
