The aim of this study is to provide a quantitative assessment of the speaker discriminating properties of broad phonetic groups. GMM based approach to speaker modelling is used in conjunction with a phonetically handlabelled speech database (TIMIT) to produce broad phonetic group ranking based on speaker identification scores. The broad phonetic groups nasals and vowels were found to be particu...

The article presents the method elaborated by the authors for automatic prolongation detection in utterances by the people who stutter. Fricative and nasals consonants were focused upon, as they are the most frequently prolonged ones. Fuzziness was applied in the scales of time, frequency and level of distinctive features of a given manner of articulation. The presented method was verified in c...

The purpose of this paper is to investigate the realization of the [nasal] contrast in French by performing an acoustic analysis of naturally spoken nasal and oral vowels, and by carrying out discriminant analysis on these data. Results consistently show that generic parametric representations allow to reliably discriminate between nasals and orals. A specific issue addressed in this paper is t...

This paper explores the possibility and extent of non-uniqueness in the acoustic-to-articulatory inversion of speech, from a statistical point of view. It proposes a technique to estimate the non-uniqueness, based on finding peaks in the conditional probability function of the articulatory space. The paper corroborates the existence of non-uniqueness in a statistical sense, especially in stop c...