Anger can be divided into two types: hot anger and cold anger. In comparison to neutral speech, cold anger is produced with a lower pitch, higher intensity, more energy (500 Hz) across the vocalization, higher first formant (first sound produced) and faster attack times at voice onset (the start of speech). Hot anger, in contrast, is produced with a higher, more varied pitch, and even greater energy (2000 Hz).[3]
Decoding emotions in speech includes three (3) stages. Determining acoustic features, creating meaningful connections of emotion with these features, and processing the acoustic patterns in relation to the connections established. In the processing stage, first system is trained for hot anger, cold anger and neutral state. During testing phase, system test sample voice and classify into angry or neutral state. …show more content…
For example, energy is a measure of emotion, but it is the fluctuation/trajectory of energy, not the energy itself makes speech sound emotional. In order to understand speech production under anger we have investigated the vocal and acoustical changes caused by the anger mental state of the speaker. Extensive evaluations on several speech production features are made. The studies have shown that the presence of anger causes changes in phoneme production with respect to glottal source factors, fundamental frequency, intensity, duration, and spectral shape [64-65]. The studies have shown that the presence of anger causes changes in phoneme production with respect to glottal source factors, fundamental frequency, intensity, duration, and spectral shape