Volume 3, Issue 1 (1-2017)                   ITCMS 2017, 3(1): 1-9 | Back to browse issues page

XML Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Kohyarnejadfard I, fathy M. Combinational gender recognition system. ITCMS. 2017; 3 (1) :1-9
URL: http://europeansp.org/article-11-170-en.html
Iran University of Science and technology
Abstract:   (66 Views)
With the rapid advances in audio processing applications such as real-time speech recognition systems, speaker recognition systems and speaker matching, existence of gender recognition system with high accuracy is very important. In this article, a system to identify the gender of speaker is proposed and also accuracy of it is shown. Proposed system has been utilized a combination of Gaussian mixture model (GMM) and Fuzzy Support Vector Machine (PSVM). At first, different characteristics of the speech signals, which are used in speaker recognition systems are described and extracted. The motivation of the use of GMM speaker modeling is presented and then the speaker’s Gaussian mixture model and Gaussian mixture density function is obtained. According to the classification power and the need to have probability of membership of an entry in a particular class, fuzzy support vector machine (PVSM) is used. Finally, results of Gaussian mixture model and fuzzy support vector machine are combined and proposed method is gained.
Full-Text [PDF 554 kb]   (12 Downloads)    

Received: 2019/08/8 | Accepted: 2019/08/8 | Published: 2019/08/8

Add your comments about this article : Your username or Email:
CAPTCHA