Emotions

The aim of the current research is to identify the main phonetic features related to the fundamental frequency changes of vocally expressed basic emotions (sadness, happiness, surprise, fear, disgust, anger) and to determine the recognition of these emotions. To achieve this, human evaluation and acoustic analysis of the phonetic properties were performed. Four professional actors’ portrayals of six emotions were presented to 42 listener judges. The results of the human perception show that the overall emotion recognition is rather good (81 proc.). The utterances were correctly identified for these emotions: sadness (96 per cent), followed by happiness (92 per cent), anger (88 per cent), surprise (85 per cent), fear (82 per cent), most errors occurred while identifying disgust (47 per cent). The latter emotion was generally confused with sadness and happiness. Anger was mostly confused with disgust, sometimes with happiness, and never with surprise. Fear and happiness were often incorrectly assigned to surprise. Four fundamental frequency parameters (mean (MF0), variation (F0SD), maximum (F0max) and range (F0 range)) were measured for the acoustic analysis. Acoustic separability was tested using a simple factorial analysis (ANOVA) and a multi-factorial analysis of variance (MANOVA) at a word level. The analysis of phonetic features revealed that similar emotions shared similar acoustic properties in all four speakers (for instance, speech associated with negative emotions was characterized by having a lower pitch, narrower F0 range and so on). Furthermore, the results of the acoustic analysis agree with the data of the perception test – the emotions are easier to recognize when their acoustic properties are most prominent: happiness and sadness were identified correctly most of the time because their phonetic features differentiated most (sadness was characterized by lowest F0 values, happiness – by the highest F0 values). Emotions that shared similar pitch values tended to be confused

Physics · Hearing Loss and Rehabilitation · Multisensory perception and integration · Noise Effects and Management