| 摘要 |
PROBLEM TO BE SOLVED: To provide a technology capable of providing information used for determining a recommendation musical piece that a user easily sings, and for accurately grading singing performance. SOLUTION: In singing pitch difference specifying processing, a pitch transition score tsc is derived (S170), by collating each of singing transition sections f0vt extracted by a singing transition section extracting processing, with a corresponding pitch transition model curve f0model (S160). These derived pitch transition scores tsc are summed up for each pitch difference section (S190). A singing possible pitch difference section ITM<SB>1</SB>in which a summed-up total transition score mtsc is a specific value or more, that is, the pitch difference section that the user can sing, is specified (S210). Then, a musical piece in which the pitch difference of a specified structure sound is within a range of the singing possible pitch difference section ITM<SB>1</SB>is proposed to the user as a recommended musical piece, and a weight which becomes larger, as the maximum pitch difference of the singing possible pitch difference section ITM<SB>1</SB>is larger, is added to a grading result (S220). COPYRIGHT: (C)2010,JPO&INPIT
|
