Comparing various voice recognition techniques

Author

Barbu, Tudor

Author_Institution

Inst. of Comput. Sci., Romanian Acad., Iasi, Romania

fYear

2009

fDate

18-21 June 2009

Firstpage

1

Lastpage

6

Abstract

In this paper we describe and compare some speaker recognition techniques. A mel-cepstral short-time analysis for speech signals is presented first. It is used in the voice feature extraction stage. Then speech-dependent and speech-independent speaker recognition approaches are proposed by us in this paper. Minimum mean-distance classification methods are used by the supervised classifier of the voice identification system. Threshold-based procedures are provided for the speaker verification part of the system.

Keywords

cepstral analysis; feature extraction; pattern classification; signal processing; speaker recognition; mel-cepstral short-time analysis; minimum mean-distance classification methods; speaker recognition techniques; speaker verification; speech signals; speech-dependent speaker recognition; speech-independent speaker recognition; supervised classifier; voice feature extraction; voice identification system; voice recognition techniques; Biometrics; Cepstral analysis; Cepstrum; Feature extraction; Loudspeakers; Mel frequency cepstral coefficient; Signal analysis; Speaker recognition; Speech analysis; Speech recognition; Hausdorff - derived metric; acoustic vectors; mel-cepstral vocal analysis; speech signal; supervised classification; text-dependent speaker recognition; text-independent speaker recognition; voice identification; voice verification;

fLanguage

English

Publisher

ieee

Conference_Titel

Speech Technology and Human-Computer Dialogue, 2009. SpeD '09. Proceedings of the 5-th Conference on

Conference_Location

Constant

Print_ISBN

978-1-4244-4727-5

Type

conf

DOI

10.1109/SPED.2009.5156172

Filename

5156172