Consonantal Recognition Using SVM and New Hierarchical Decision Structure Based in the Articulatory Phonetics

Author

de Andrade Bresolin, A. ; Neto, Adrião Duarte D ; Alsina, Pablo Javier

Author_Institution

Technol. Fed. Univ. of the Parana, Curitiba, Brazil

fYear

2008

fDate

15-17 Dec. 2008

Firstpage

545

Lastpage

550

Abstract

In this work, a new concept of making the consonantal recognition is proposed. We used several units (phonemes, diphones and syllables) to make the word recognition. This concept was carried out by a new hierarchical decision structure, based on the articulatory phonetics and SVM (support vector machine). The speech features used were MFCC (mel-frequency cepstral coefficient) and WPT (waveletpacket transform). 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.

Keywords

cepstral analysis; natural languages; speech processing; speech recognition; support vector machines; wavelet transforms; Brazilian Portuguese language; MFCC; SVM; WPT; articulatory phonetics; consonantal recognition; hierarchical decision structure; mel-frequency cepstral coefficient; speech features; support vector machine; waveletpacket transform; word recognition; Cepstral analysis; Mel frequency cepstral coefficient; Spatial databases; Speech recognition; Support vector machine classification; Support vector machines; Testing; Tongue; Wavelet packets; Wavelet transforms; SVM- Support Vector Machine; Speech Recognition; Wavelet Packet Transform;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimedia, 2008. ISM 2008. Tenth IEEE International Symposium on

Conference_Location

Berkeley, CA

Print_ISBN

978-0-7695-3454-1

Electronic_ISBN

978-0-7695-3454-1

Type

conf

DOI

10.1109/ISM.2008.61

Filename

4741224