| 摘要 |
<P>PROBLEM TO BE SOLVED: To reduce a load involved in defining relationship between inputs and tone data. <P>SOLUTION: A storage section 30 stores a fragment data DS of each of fragments segmented from a music piece. The fragment data DS includes a tone data M of the fragment and a descriptor P indicative of a musical character of the fragment. A descriptor generation section 12 generates a descriptor Q indicative of a musical character from input data I based on operation on a musical performance device 60. A similarity determination section 14 determines a similarity index value R indicative of similarity between the descriptor Q and the descriptor P of each of the fragments. A fragment selection section 16 selects the fragment based on the similarity index value R of each of the fragments. On the basis of the tone data M of each of the fragments selected by the fragment selection section 16, a data generation section 18 generates an output data O. <P>COPYRIGHT: (C)2008,JPO&INPIT |
