DocumentCode
3540363
Title
PNN-based algorithm for the recognition of speakers
Author
Fang, Ye ; Zhou, Yabin
Author_Institution
Sch. of Electron. Inf., Xi´´an Polytech. Univ., Xi´´an, China
fYear
2009
fDate
16-19 Aug. 2009
Abstract
Based on analysis of the conventional speech signal identical algorithm, an improved algorithm for the speaker recognition is introduced. It is a method based on the probabilistic neural network (PNN), which uses Mel Frequency Ceptral Coefficients(MFCC) as human sound feature parameters. The experiment has shown the high accuracy of the proposed algorithm in the classification of training samples and testing samples. The experiment results also validate the effectiveness of the method.
Keywords
neural nets; probability; speaker recognition; human sound feature; mel frequency ceptral coefficient; probabilistic neural network; speaker recognition; speech signal identical algorithm; Algorithm design and analysis; Classification algorithms; Frequency; Humans; Loudspeakers; Neural networks; Signal analysis; Speaker recognition; Speech analysis; Testing; Feature abstraction; MFCCs; PNN; Speaker recognition; Speech identification;
fLanguage
English
Publisher
ieee
Conference_Titel
Electronic Measurement & Instruments, 2009. ICEMI '09. 9th International Conference on
Conference_Location
Beijing
Print_ISBN
978-1-4244-3863-1
Electronic_ISBN
978-1-4244-3864-8
Type
conf
DOI
10.1109/ICEMI.2009.5273997
Filename
5273997
Link To Document