| 摘要 |
PROBLEM TO BE SOLVED: To provide the voice inputting device and the method and the storage medium through which adequate intervals of the first voice portion of the voice to be pronounced are easily recognized beforehand. SOLUTION: In a musical sound data generating mode, a user arbitrary sets double sounds such as the main chord of the key of a melody or the single sound such as a root sound as arbitrary reference sounds before inputting the final sounds (a step S403). Then, musical sound data are obtained by pitch detection and musical sound assignment based on the voice information related to a portion of the melody (for the amount of few bars), detections are made for the chord relative to the musical sound data or the key and the chord of the start of a singing of the first bar, for example or the key of the main chord are automatically set as the reference sounds (a step S404). Then, the set reference sounds are generated immediately prior to the input start of the final sounds (a step S407). |
