Adult Speech in Different Emotional States: Temporal and Spectral Features

Abstract

The aim of the study is to determine individual features of adult speech in different emotional states. The acoustic speech characteristics of 12 adult native Russian speakers were studied. The speech of informants uttering meaningless phrase in different emotional states was audio recorded: joy, anger, sadness, fear, and neutral. The temporal and spectral characteristics of speech were analyzed in the Cool Edit Pro sound editor. The maximum pitch range in male speech is revealed in phrases uttered in a neutral state and a state of joy; the minimum, in a state of sadness. For female speech, the maximum pitch range is in a state of joy and in a state of anger; the minimum, in a state of sadness and in a neutral state. The pitch range in female speech is larger than that in male speech. For seven informants, it was shown that the duration of utterances in a state of sadness was longer compared to other states, and in a state of joy, on the contrary, it was minimal. Both male and female utterances in a state of joy were characterized by maximum pitch range values; conversely, in a state of sadness, by minmum values. Pauses between words in utterances in a state of sadness were detected in both men and women. Thus, differences in the temporal and spectral characteristics of utterances in different emotional states are revealed. The individual features of the manifestation of the emotional state in the speech of adults are determined.