LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION

LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION

Speaker Identification (SI) aims at automatically identifying an individual by extracting and processing information from his/her voice. Speaker voice is a robust a biometric modality that has a strong impact in several application areas. In this study, a new combination learning scheme has been pro...

Full description

Bibliographic Details
Main Authors: IMEN TRABELSI, MED SALIM BOUHLEL
Format: Article
Language:English
Published: Taylor's University 2017-05-01
Series:Journal of Engineering Science and Technology
Subjects:
Speaker identification
LVQ
GMM
MFCC
TIMIT
Online Access:http://jestec.taylors.edu.my/Vol%2012%20issue%205%20May%202017/12_5_1.pdf
id doaj-3bbb2477917d4d4e9d47492b4e25eab9
record_format Article
spelling doaj-3bbb2477917d4d4e9d47492b4e25eab92020-11-24T23:40:59ZengTaylor's UniversityJournal of Engineering Science and Technology1823-46902017-05-0112511531164LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATIONIMEN TRABELSI0MED SALIM BOUHLEL1Sciences and Technologies of image and Telecommunications (SETIT), University of Sfax, TunisiaSciences and Technologies of image and Telecommunications (SETIT), University of Sfax, TunisiaSpeaker Identification (SI) aims at automatically identifying an individual by extracting and processing information from his/her voice. Speaker voice is a robust a biometric modality that has a strong impact in several application areas. In this study, a new combination learning scheme has been proposed based on Gaussian mixture model-universal background model (GMM-UBM) and Learning vector quantization (LVQ) for automatic text-independent speaker identification. Features vectors, constituted by the Mel Frequency Cepstral Coefficients (MFCC) extracted from the speech signal are used to train the New England subset of the TIMIT database. The best results obtained (90% for gender- independent speaker identification, 97 % for male speakers and 93% for female speakers) for test data using 36 MFCC features. http://jestec.taylors.edu.my/Vol%2012%20issue%205%20May%202017/12_5_1.pdfSpeaker identificationLVQGMMMFCCTIMIT
collection DOAJ
language English
format Article
sources DOAJ
author IMEN TRABELSI
MED SALIM BOUHLEL
spellingShingle IMEN TRABELSI
MED SALIM BOUHLEL
LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION
Journal of Engineering Science and Technology
Speaker identification
LVQ
GMM
MFCC
TIMIT
author_facet IMEN TRABELSI
MED SALIM BOUHLEL
author_sort IMEN TRABELSI
title LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION
title_short LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION
title_full LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION
title_fullStr LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION
title_full_unstemmed LEARNING VECTOR QUANTIZATION FOR ADAPTED GAUSSIAN MIXTURE MODELS IN AUTOMATIC SPEAKER IDENTIFICATION
title_sort learning vector quantization for adapted gaussian mixture models in automatic speaker identification
publisher Taylor's University
series Journal of Engineering Science and Technology
issn 1823-4690
publishDate 2017-05-01
description Speaker Identification (SI) aims at automatically identifying an individual by extracting and processing information from his/her voice. Speaker voice is a robust a biometric modality that has a strong impact in several application areas. In this study, a new combination learning scheme has been proposed based on Gaussian mixture model-universal background model (GMM-UBM) and Learning vector quantization (LVQ) for automatic text-independent speaker identification. Features vectors, constituted by the Mel Frequency Cepstral Coefficients (MFCC) extracted from the speech signal are used to train the New England subset of the TIMIT database. The best results obtained (90% for gender- independent speaker identification, 97 % for male speakers and 93% for female speakers) for test data using 36 MFCC features.
topic Speaker identification
LVQ
GMM
MFCC
TIMIT
url http://jestec.taylors.edu.my/Vol%2012%20issue%205%20May%202017/12_5_1.pdf
work_keys_str_mv AT imentrabelsi learningvectorquantizationforadaptedgaussianmixturemodelsinautomaticspeakeridentification
AT medsalimbouhlel learningvectorquantizationforadaptedgaussianmixturemodelsinautomaticspeakeridentification
_version_ 1725508436265795584

Similar Items