作者: Joe Crumpton , Cindy L. Bethel
关键词:
摘要: This research investigated the use of Mary, an open source speech synthesizer, to convey emotional intent a robot through robot's vocal prosody. The pitch, pitch range, rate, and volume were varied anger, fear, happiness, sadness. results indicate participants recognized intended emotions at rate statistically higher than chance whether they provided their own free choice word or used forced survey describe emotion presented by robot. As expected, correctly transcribed more words spoken when prosody modifications performed small (sadness) large (anger fear).