Consonantal recognition using SVM and a hierarchical decision structure based in the articulatory phonetics

Abstract

A new concept of making the Consonantal Recognition is proposed in this work, where used units (phonemes and syllables) to make the word recognition. This concept was carried out by a hierarchical decision structure, based on the Articulatory Phonetics and SVM. The speech features used were MFCC and WPT. Eighteen consonantal phonemes have been used in the recognition. The database used for the recognition was a set of two-syllable words of the Brazilian Portuguese language. The experimental results showed success rates of 98.41% for the user-dependent case. Our focus was the dependent speaker in order to validate the new proposal. © 2012 Springer-Verlag.