| 摘要 |
<P>PROBLEM TO BE SOLVED: To automatically compose an appropriate musical piece suitable for lyrics, even if a user does not enter a parameter which is other than the lyrics. <P>SOLUTION: An arithmetic model is learned, by using as teacher data, a language feature quantity calculated from lyrics data showing lyrics of each of a plurality of musical pieces and attribute data that show the attribute of the musical piece. The information processing apparatus includes a storage section for storing the arithmetic model for deriving the attribute data of the musical piece from the language feature quantity of the lyrics data; a character string acquiring section for acquiring a character string input from the user; a feature quantity calculating section for calculating the language feature quantity from the acquired character string; a model arithmetic section for deriving the attribute data of the musical piece to be composed from the language feature quantity calculated by the feature quantity calculation section according to the arithmetic model; and a composition section for forming a musical piece where the character string acquired by the character string acquiring section is used as the lyrics, by using the attribute data derived by the model arithmetic section. <P>COPYRIGHT: (C)2011,JPO&INPIT |
