This paper presents a technique using artificial neural networks (ANNs) for speaker identification that results in a better success rate compared to other techniques. The technique used in this paper uses both power spectral densities (PSDs) and linear prediction coefficients (LPCs) as feature inputs to a self organizing feature map to achieve a better identification performance. Results for speaker identification with different methods are presented and compared.
V. Moonasar and G. K. Venayagamoorthy, "Speaker Identification Using a Combination of Different Parameters as Feature Inputs to an Artificial Neural Network Classifier," Proceedings of IEEE Africon, 1999, Institute of Electrical and Electronics Engineers (IEEE), Jan 1999.
The definitive version is available at http://dx.doi.org/10.1109/AFRCON.1999.820791
IEEE Africon, 1999
Electrical and Computer Engineering
Keywords and Phrases
ANN; Artificial Neural Network Classifier; Feature Inputs; Identification Performance; Linear Prediction Coefficients; Pattern Classification; Power Spectral Densities; Prediction Theory; Self Organizing Feature Map; Self-Organising Feature Maps; Speaker Identification; Speaker Recognition; Spectral Analysis; Success Rate
Article - Conference proceedings
© 1999 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.