https://doi.org/10.31174/SEND-Ph2020-240VIII71-09 

Annotation. The article considers the problems of determining  the emotional states of the speaker with the help of linguistic and paralinguistic means. As we speak, we convey to our listeners information about our emotional state and attitudes toward both our listeners and what we say. Although in general the specific ability to decode the emotional component of the utterance depends on the individual emotion.It is known that native speakers, when listening to actors who read an emotionally neutral text, while depicting emotions, perceptually are not always able to correctly identify them. Even if the emotional expression through prosody may not always be consciously recognized by the participants in the conversation, it may be subconsciously influenced by other indicators, such as the tone of voice, for example, or facial expressions. This type of expression follows not only from linguistic or semantic effects. The ability of an average person to decipher emotional states during conversations is less developed than the traditional ability to identify facial expressions. Whether a person perceives prosody as positive, negative or neutral, he always coordinates  with facial expressions. If it becomes closer to neutral, the interpretation of the emotional state is influenced by prosodic interpretation. To decide on the level of emotionality, listeners are often forced to listen to the prosody of the statement. To be able to determine the level of affectivity of the statement, listeners need at least 600 milliseconds of prosodic information, below which there is not enough time to process the emotional context of the statement. As previous research has focused on such prosodic features of speech as basic average frequency and range, tempo and intensity, it is now necessary to analyze data using a wider range of voice effects, including paralinguistic. Such studies have not been conducted in the past, although they play an important role in the depiction of emotions. Having developed a scale with a gradient distinction between prosodic and paralinguistic features in speech, researchers placed into- nation with its indicators, which indicates its linguistic character, at the prosodic end of the scale, and at the other end of the scale – such features as voice quality, or noise and all sorts of clicks as  paralinguistic features. Nonlinguistic or paralinguistic features are classified  into  two  types:  individual  variations  and  reflexes.  Individual  variation  includes  the  effects  of  phonation  and  resonance through the physiology of the speaker and the features of his vocal tract. Reflex states should also be considered, as they are often an involuntary sign of true emotional stress. In general, it should be noted that emotional or affective prosody is manifested in both lin- guistic and paralinguistic aspects of speech, which allow and help both to convey and understand emotions. It includes an individual tone of voice, which is transmitted through changes in pitch, volume, timbre, speech rate and pauses.