| 摘要 |
PURPOSE:To obtain a reproduced voice of high quality by generating a reproduced voice signal from a generated sound source parameter and a spectrum envelope parameter sent from an encoding side. CONSTITUTION:The voice decoder has a code input terminal 310 to which a code sequence is inputted and is equipped with a demultiplexer 315, a linear predictive decoder 320, a pitch predictive decoder 325, a sound source parameter decoder 330, a synthesizing filter reproducer 340, a sound source signal reproducer 350, and a reproduced voice signal generator 370, and a reproduced voice signal is outputted from an output terminal 380. Then a sound source signal with constant internal length is generated from a basic period sent from a transmission side and a sound source parameter in one basic period section and the reproduced voice signal is generated from the generated sound source signal and the spectrum envelope parameter sent from an encoding side to obtain a reproduced voice. Consequently, the reproduced voice signal of high quality is obtained. |
