期刊名称：International Journal of Computer Science & Information Technology (IJCSIT)
印刷版ISSN：0975-4660
电子版ISSN：0975-3826
出版年度：2016
卷号：8
期号：2
页码：63
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：In the era of data explosion, speech emotion plays crucial commercial significance. Emotion recognition inspeech encompasses a gamut of techniques starting from mechanical recording of audio signal to complexmodeling of extracted patterns. Most challenging part of this research purview is to classify the emotion ofthe speech purely based on the physical characteristics of the audio signal independent of language ofspeech. This paper focuses on the predictive modeling of audio speech data based on most viable featureset extraction and deployment of these features to predict the emotion of unknown speech data. We haveused two most widely used classifiers, a variant of CART and Naïve Bayes, to model the dynamics ofinterplay of crucial features like Root Mean Square (RMS), Zero Cross Rate (ZCR), Pitch and Brightness ofaudio signal to determine the emotion of speech. In order to carry out comparative analysis of the proposedclassifiers, a set of experiments on real speech data is conducted. Results clearly indicate that decision treebased classifier works well on accuracy whereas Naïve Bayes works fairly well on generality.
关键词：Acoustic features; audio emotion recognition; speech emotions and predictive classifier.
