| 摘要 |
PROBLEM TO BE SOLVED: To minimize deviation between an evaluation result by an apparatus and an evaluation by a listener without preparing scoring reference data for every music piece in advance in the apparatus for evaluating a skill of singing and a musical performance.SOLUTION: A scoring unit 12 identifies a pitch, sound volume, and a voice quality of singing voice data as features, and evaluates the singing voice data on a plurality of evaluation items such as "music interval", "rhythm", and "modulation" based on the identified features. The scoring unit 12 identifies features (a progress of code, key, rhythm, and the like) of accompaniment sound from accompaniment sound data to identify a value of a weighting coefficient for every evaluation item on the basis of the identified features. The scoring unit 12 weights an evaluation value for every evaluation item using the identified weighting coefficient for every evaluation item and evaluates the singing voice data according to the weighting results. |
