摘要 |
PROBLEM TO BE SOLVED: To evaluate whether or not a singer correctly remembers lyrics without making a system complicated. SOLUTION: A CPU 102 divides the voice waveform of a model voice that model voice data represent into a plurality of frames. Further, the CPU 102 divides the voice waveform of a singing voice that singing voice data stored in a storage section 105 represent into a plurality of frames. Further, the CPU 102 makes voice waveforms of the respective frames of the model voice correspond to voice waveforms of the respective frames of the singing voice. The CPU 102 compares formant frequencies of the voice waveforms of the corresponding frames with each other to judge whether or not the model voice and singing voice match each other. When it is judged that the model voice and singing voice do not match each other, lyrics corresponding to a voice that the model voice data represent is displayed on a monitor 2. COPYRIGHT: (C)2007,JPO&INPIT
|