Аннотация:Our main hypothesis would be as follows: a) temporal structuring of speech is very individual and most dependent on the phonetic-phonological material of a language; b) temporal structuring of speech is most important for detection of changes in emotional state of a speaker (especially for fear, anxiety). In this case all groups of basic emotions conclude the emotion of fear. From this point of view we can define it as a specific universal emoteme (as phonemic sign) concerning cognition level of theory of emotions. Long experience in carrying out real forensic examinations shows that the main difficulty in this field consists in the procedures of choosing parameters which can help us to identify the emotional state of a suspect. In this case the perceptual-auditory and acoustic analysis of the speech material is very important. The overall effect of temporal parameter variations can significantly raise the performance of speech/speaker recognition systems.