| 摘要 |
PURPOSE: To attain a polyphonic reproduction by extracting the pitch of a voice, outputting the voice as it is when the voice is judged as a voiceless sound, and at the time of judging the voice as a voiced sound, compressing the voice waveform of one extracted pitch and repeatedly outputting the waveform of one pitch. CONSTITUTION: A pitch extraction circuit 8 extracts one pitch from an inputted voice waveform. The voice waveform of one pitch is compressed or extended about a time base by a sound level inputted to a sound level input circuit. Namely in the case of reproducing a voice higher than a voice inputted to an input terminal 1 by one octave, a time base converting/repeating circuit 9 compresses the voice waveform of one pitch into a half about the time base and repeats the waveform of one pitch twice. When a voice inputted to a voiceless/voiced sound discriminating circuit 6 is a voiceless sound, a switching circuit 7 outputs an output signal from a delay circuit 5 as it is because the voiceless sound is independent of the height of the voice. Consequently, the polyphonic performance using voices can be attained. |
