DocumentCode :
2863689
Title :
Comparison of linear predictive analysis methods for ANN-based speaker identification
Author :
Inal, M. ; Bütün, Erhan ; Erkan, Kadir ; Yildirum, Mehmet ; Çeken, Celal
Author_Institution :
Dept. of Electron. & Comput. Educ., Kocaeli Univ., Japan
fYear :
2000
fDate :
2000
Firstpage :
109
Lastpage :
112
Abstract :
This paper studies the parametrization of speech using linear predictive (LP) analysis and its application to speaker identification. The task of speaker classification is realized using multilayer perceptron. The system is evaluated on speaker set for speaker identification. The system is also tested with cepstral and reflection coefficients, and the results obtained are compared
Keywords :
feature extraction; linear predictive coding; multilayer perceptrons; speaker recognition; cepstral coefficient; feature extraction; linear predictive model; multilayer perceptron; neural nets; reflection coefficient; speaker identification; Autoregressive processes; Cepstral analysis; Computer science education; Feature extraction; Multilayer perceptrons; Predictive models; Signal analysis; Speaker recognition; Speech analysis; Time series analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Network Applications in Electrical Engineering, 2000. NEUREL 2000. Proceedings of the 5th Seminar on
Conference_Location :
Belgrade
Print_ISBN :
0-7803-5512-1
Type :
conf
DOI :
10.1109/NEUREL.2000.902395
Filename :
902395
Link To Document :
بازگشت