Automatic Speech Emotion Recognition Using Support Vector Machine

摘要：

Automatic Speech Emotion Recognition (SER) is a current research topic in the field of Human Computer Interaction (HCI) with wide range of applications. The purpose of speech emotion recognition system is to automatically classify speakers utterances into five emotional states such as disgust, boredom, sadness, neutral, and happiness. The speech samples are from Berlin emotional database and the features extracted from these utterances are energy, pitch, linear prediction ccpstrum coefficients (LPCC), Mel Frequency cepstrum coefficients (MFCC), Linear Prediction coefficients and Mel cepstrum coefficients (LPCMCC). The Support Vector Machine (SVM) is used as a classifier to classify different emotional states. The system gives 66.02％ classification accuracy for only using energy and pitch features, 70.7％ for only using LPCMCC features, and 82.5％ for using both of them.

关键词： Speech Emotion Automatic Emotion Recognition SVM Energy Pitch LPCC MFCC LPCMCC

作者: Peipei Shen Zhou Changjun Xiong Chen

作者单位: Department of Computer Technology Shanghai Jiao Tong University Shanghai, China Pudong Branch China Mobile Group Shanghai Company Limited Shanghai, China

会议类型: 国际会议

会议名称: 2011 International Conference on Electronic & Mechanical Engineering and Information Technology(EMEIT 2011)(2011年机电工程与信息技术国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 621-625

在线出版日期: 2011-08-12（万方平台首次上网日期，不代表论文的发表时间）