The role of voice fundamental frequency in the perception of anger in clear speech

Abstract

In previous work, listeners rated clearly spoken neutral speech materials as sounding angry significantly more often than identical materials spoken conversationally. This effect varies widely among talkers, and a recent study compared acoustic characteristics of clear and conversational speech produced by talkers whose clear speech sounds angry to those of talkers whose clear speech does not sound angry. Of several well-known clear speech acoustic changes, only raised voice fundamental frequency (fo) differed for talkers who do and do not sound angry when speaking clearly. To test whether raised fo causes clear speech to sound angry, the present study used six talkers whose clear speech was rated as sounding angry and who also made sizeable fo shifts when speaking clearly. For each talker, nine pairs of identical sentences (one from each speaking style) were identified. Praat was then used to lower the pitch of the clear sentence to match the pitch of the conversational sentence and to raise the pitch of the conversational sentence to match the pitch of the clear sentence. Listeners with normal hearing will hear these sentences and rate the emotion they heard in each item from a set of six emotions.