Machines imitating humans appearance and behaviour in robots

Abstract

The authors have synthesized the emotion in the speech of robot. The modeling of emotion in speech relies on a number of parameters among others, fundamental frequency (F0) level, voice quality, or articulation precision etc. As an initial work for synthesizing emotion in speech, we utilized the three voice features provided by the TTS engine of Microsoft Speech SDK i.e. pitch, rate and volume. Speech with these parameters controlled, was generated randomly with 20 sentences for each emotion and perception by human hearers were collected. © Springer Science+Business Media B.V. 2009.