Feature selection and dimension reduction for automatic gender identification

Keyvanrad, Mohammad Ali; Homayounpour, Mohammad Mehdi

doi:10.1109/csicc.2009.5349647

Cited by 5 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two features including MFCC 6 with 21 coefficients as spectral features and pitch as a prosodic feature are used. Higher MFCC coefficients are used since in some feature selection experiments using SOAP, it was observed that the higher MFCC order coefficients are usually selected as best coefficients for gender identification [13]. Also hierarchical model and classifiers in each level are obtained using training and development data (described in section II).…”

Section: Hierarchical Classification For Automatic Gender Identifimentioning

confidence: 99%

Performance improvement in automatic gender identification using hierarchical clustering

Keyvanrad

Homayounpour

2010

2010 5th International Symposium on Telecommunications

Self Cite

View full text Add to dashboard Cite

In this paper a hierarchical structure is proposed for automatic gender identification (AGI). In this structure two clustering techniques are used. The first technique is divisive clustering for dividing speakers from each gender to some classes of speakers. The second clustering technique is agglomerative clustering for creating a hierarchical structure. Feature reduction is done by SOAP feature selection to reduce clustering and classification run time. Another idea used in this paper is to choose and use the best classifier in each level of the proposed hierarchical gender classifier. Three classifiers including GMM, SVM and MLP are used. 96.19% gender classification accuracy was obtained on OGI Multilanguage Telephony Corpus which is 3.5%, 2.5%, and 1% better than the performance obtained by traditional classifiers using GMM, SVM and MLP respectively.

show abstract

Section: Hierarchical Classification For Automatic Gender Identifimentioning

confidence: 99%

Performance improvement in automatic gender identification using hierarchical clustering

Keyvanrad

Homayounpour

2010

2010 5th International Symposium on Telecommunications

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most previous studies on gender classification focused only on the speaker voice characteristics of adults and achieved high levels of accuracy. 21 However, gender classification in children still remains a challenge because of some difficulties in distinguishing between the fundamental and formant frequencies of male and female children.…”

Section: Introductionmentioning

confidence: 99%