Title :
Voice verification based on nonlinear Ho-Kashyap classifier
Author_Institution :
Silesian Univ. of Technol., Gliwice
Abstract :
This article presents enhancement of a linear Ho-Kashyap classifier based on kernel functions applied to speaker verification. Obtained decision function is nonlinear. Achieved verification accuracy is compared with classical techniques like Gaussian mixture models and vector quantization. Polish speech corpora ROBOT was applied as a training and testing set. Obtained results are discussed.
Keywords :
pattern classification; speech recognition; Gaussian mixture models; Polish speech corpora ROBOT; kernel functions; nonlinear Ho-Kashyap classifier; speaker verification; vector quantization; voice verification; Cepstral analysis; Data acquisition; Feature extraction; Multidimensional systems; Pattern matching; Robotics and automation; Speaker recognition; Speech; Telephony; Testing;
Conference_Titel :
Computational Technologies in Electrical and Electronics Engineering, 2008. SIBIRCON 2008. IEEE Region 8 International Conference on
Conference_Location :
Novosibirsk
Print_ISBN :
978-1-4244-2133-6
Electronic_ISBN :
978-1-4244-2134-3
DOI :
10.1109/SIBIRCON.2008.4602653
