| 摘要 |
<P>PROBLEM TO BE SOLVED: To make a robot or the like adapt self action to stimuli given by others. <P>SOLUTION: In an information processing apparatus, a recognition part 5 recognizes utterance of a user by an auditory sensation SOM stored in a storage part 4, and a generation part 6 changes articulation parameters which are obtained from an articulation SOM stored in the storage part 4 in order to output a synthetic sound obtained by imitating the utterance, and a driving part 8 actually outputs the synthetic sound by performing speech synthesis according to the articulation parameters after the change. Moreover, the recognition part 5 recognizes the synthetic sound, and decides whether the synthetic sound is close to the utterance of the user or not. When the synthetic sound is close to the utterance of the user, a learning part 7 strengthens the coupling between the node of the articulation SOM which can generate the synthetic sound close to the utterance of the user and the node of the auditory sensation SOM which is the recognition result of the synthetic sound or the utterance of the user. This invention is applicable to, for example, the robot or the like. <P>COPYRIGHT: (C)2006,JPO&NCIPI |
