Estimation of articulatory synthesiser parameters from pseudo-formants

Abstract

The articulatory speech synthesiser is likely to be the ultimate solution to the synthesis of natural sounding, intelligible speech. Yet, the problem of estimating articulatory parameters, from a given speech signal, remains a challenge although remarkable attempts have been reported within the literature towards this end. This paper presents a new technique for the accurate estimation of articulatory parameters through the use of "pseudo formants" and their corresponding amplitudes.