6. Analysis of Head Motions and Speech, and Head Motion Control in an Android Robot

With the aim of automatically generating head motions during speech utterances, analyses are conducted to verify the relations between head motions and linguistic and paralinguistic information carried by speech utterances. Motion-captured data are recorded during natural dialogue, and the rotation angles are estimated from the head marker data. Analysis results show that nods frequently occur during speech utterances, not only for expressing specific dialogue acts such as agreement and affirmation, but also to indicate syntactic or semantic units, which appear at the last syllable of the phrases, in strong phrase boundaries. The dependence on linguistic, prosodic and voice quality information of other head motions, including shakes and tilts, is also analyzed, and the potential for using this to automatically generate head motions is discussed. Intra-speaker variability and inter-speaker variability on the relations between head motion and dialogue acts are also analyzed. Finally, a method for controlling the head actuators of an android based on the rotation angles is proposed, and the mapping from human head motions is evaluated.