Speech conversion and its application to alaryngeal speech enhancement
Ning Bi, Yingyong Qi · 2002
In this work, two speech conversion algorithms, a vector quantization (VQ) based algorithm and a linear multivariate regression (LMR) based algorithm, were modified. The modifications were aimed at reducing the spectral distortion (bandwidth increase) in the VQ-based algorithm and the spectral discontinuity in the LMR-based algorithm. The modified algorithms were used for the enhancement of alaryngeal speech. Results of perceptual evaluations indicated that listeners generally preferred to listen to the enhanced speech samples.