AN ADAPTIVE POST-FILTERING TECHNIQUE BASED ON THE MODIFIED YULE-WALKER FILTER

摘要

An adaptive time domain post-filtering technique based on the Yul e-Walker filter is presented for designing a filter for filtering a speech signal. Wherein the desired frequency response is specified, the denominator coefficients are determined according to a least squares approach (106) based on non-recursive correlation coefficients computed by inverse Fourier Transformation (IFFT) of the specified frequency response, the numerator polynomial is determined by additive decomposition (108). Spectral factorization is applied (110) to enable the impulse response to be calculated (112) and the method of least squares is used to determine the f inal denominator polynomial (114). Information is gathered about the relation between poles and formants and then the formant and their bandwidth are estimated. The information about the formants and their bandwidth is then used to design the modified Yule-Walker filter based on a least squares fit in the time domain.