Method and apparatus for determining voiced / unvoiced sound and method for encoding voice

摘要

An apparatus and method for voiced / unvoiced discrimination for determining whether an input speech signal is voiced or unvoiced. The parameters for discriminating voiced / unvoiced (V / UV) are extensively determined in order to perform V / UV discrimination with high accuracy by a simplified algorithm. (V / UV) discrimination parameter for the input speech signal, the frame average energy lev of the input speech signal, the normalized autocorrelation peak value r0r, the spectral similarity pos, the eigenvalue nzero, And a pitch lag (pch) is supplied to the input terminals 11 to 15. If these parameters are represented by x, these parameters are converted by the function calculation circuits 31 to 35 by the sigmoid function g (x) expressed by the following equation.;g (x) = A / (1 + exp (-(x-b) / a);Where A, a, and b are constants. By using the parameters converted by the sigmoid function g (x), the voiced / unvoiced sound is discriminated by the V / UV discrimination circuit 26.