Modeling emotion and attitude in speech by means of perceptually based parameter values

This study focuses on the perception of emotion and attitude in speech. The ability to identify vocal expressions of emotion and/or attitude in speech material was investigated. Systematic perception experiments were carried out to determine optimal values for the acoustic parameters: pitch level, p itch range and speech rate. Speech was manipulated by varying these paramet ers around the values found in a selected subset of the speech material whi ch consisted of two sentences spoken by a male speaker expressing seven emo tions or attitudes: neutrality, joy, boredom, anger, sadness, fear, and ind ignation. Listening tests were carried out with this speech material, and o ptimal values for pitch level, pitch range, and speech rate were derived fo r the generation of speech expressing emotion or attitude, from a neutral u tterance. These values were perceptually tested in re-synthesized speech an d in synthetic speech generated from LPC-coded diphones.