Fuzzy logic and neural network based gender classification using three features
by K. Meena; K.R. Subramaniam; M. Gomathy
International Journal of Signal and Imaging Systems Engineering (IJSISE), Vol. 7, No. 2, 2014

Abstract: Gender classification is one of the most important processes in speech processing. Generally gender classification is done by considering pitch as feature. Normally the pitch value of female is higher than the male. By using this condition, gender classification process takes place. But in some case the pitch value of male is higher and also pitch of female is low, in that case this classification does not provide the exact result. By considering the abovementioned drawback, here proposed a new method for gender classification in speech processing which considers three features and uses fuzzy logic and neural network to identify the gender of the speaker. The features considered in the proposed method is energy entropy, short time energy and zero crossing rate. For training fuzzy logic and neural network, training dataset is generated using the above three features. Then mean value is computed from the result obtained from fuzzy logic and neural network. The gender classification is done by using this mean value. The implementation result shows the performance of the proposed technique in gender classification.

Online publication date: Mon, 27-Oct-2014

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Signal and Imaging Systems Engineering (IJSISE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com