Improved $F_{0}$ Estimation based on $\ell_{2}$-norm Regularized TV-CAR Analysis using Bone-Conducted filter with an Adaptive Pre-Emphasis *

Abstract

Linear Prediction (LP) used in the CELP speech coding is the most successful speech analysis method. However, it has drawbacks, such as low estimation accuracy due to its least-squares scheme. We proposed time-varying complex AR (TV-CAR) speech analysis based on MMSE criterion, robust criterion, and LASSO analysis and evaluated it for $F_{0}$ estimation of speech. We also proposed $\ell_{2}$-norm regularized TV-CAR analysis, Regularized LP (RLP)-based, Time-RLP (TRLP)-based and their combined methods, and evaluated them on $F_{0}$ estimation based on the IRAPT using complex residual signals. In addition, a bone-conducted (BC) pre-filter has already been introduced to improve performance. This paper proposes an improved $F_{0}$ estimation lntroducing adaptive pre-emphasis and shows its effectiveness.