Methods for stress classification: nonlinear TEO and linear speech based features

Speech production variations due to perceptually induced stress contribute signiicantly to reduced speech processing performance. One approach that can improve the robustness of speech processing (e.g., recognition) algorithms against stress is to formulate an objective classiication of speaker stress based upon the acoustic speech signal. In this paper , an overview of recent methods for stress classiication is presented. First, we review traditional pitch-based methods for stress detection and classiication. Second, neural network based stress classiiers with cepstral-based features, as well as wavelet-based classiication algorithms are considered. The eeect of stress on linear speech features is discussed , followed by the application of linear features and the Teager Energy Operator (TEO) based nonlinear features for eeective stress classiication. A new evaluation for stress classiication and assessment is presented using a critical band frequency partition based TEO feature and the combination of several linear features. Results using Nato databases of actual speech under stress are presented. Finally , we discuss issues relating to stress classiication across known and unknown speakers and suggest areas for further research.