Open AccessUsing Neural Network with Speaker Applications
Author(s) -
Baghdad Science Journal
Publication year - 2010
Publication title -
mağallaẗ baġdād li-l-ʿulūm
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.167
H-Index - 6
eISSN - 2411-7986
pISSN - 2078-8665
DOI - 10.21123/bsj.7.2.1076-1081
Subject(s) - speech recognition , computer science , preprocessor , timit , identification (biology) , multilayer perceptron , pattern recognition (psychology) , artificial neural network , speaker identification , feature (linguistics) , perceptron , projection (relational algebra) , artificial intelligence , speaker recognition , set (abstract data type) , noise (video) , hidden markov model , algorithm , linguistics , philosophy , botany , image (mathematics) , biology , programming language
In Automatic Speech Recognition (ASR) the non-linear data projection provided by a one hidden layer Multilayer Perceptron (MLP), trained to recognize phonemes, and has previous experiments to provide feature enhancement substantially increased ASR performance, especially in noise. Previous attempts to apply an analogous approach to speaker identification have not succeeded in improving performance, except by combining MLP processed features with other features. We present test results for the TIMIT database which show that the advantage of MLP preprocessing for open set speaker identification increases with the number of speakers used to train the MLP and that improved identification is obtained as this number increases beyond sixty. We also present a method for selecting the speakers used for MLP training which further improves identification performance.