Presenting a New Text-Independent Speaker Verification System Based on Multi Model GMM

Publish Year: 1393
نوع سند: مقاله ژورنالی
زبان: English
View: 624

This Paper With 12 Page And PDF Format Ready To Download

  • Certificate
  • من نویسنده این مقاله هستم

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این Paper:

شناسه ملی سند علمی:

JR_JACR-5-4_007

تاریخ نمایه سازی: 16 شهریور 1395

Abstract:

Speaker verification is the process of accepting or rejecting claimed identity interms of its sound features. A speaker verification system can be used for numeroussecurity systems, including bank account accessing, getting to security points,criminology and etc. When a speaker verification system wants to check the identityof individuals remotely, it confronts problems such as noise effect on speech signaland also identity falsification with speech synthesis. In this system, we haveproposed a new speaker verification system based on Multi Model GMM, called SVMMGMM,in which all speakers are divided into seven different age groups, andthen an isolated GMM model for each group is created; instead of one model for allspeakers. In order to evaluate, the proposed method has been compared with severalspeaker verification systems based on Naïve, SVM, Random Forest, Ensemble andbasic GMM. Experimental results show that the proposed method has so betterefficiency than others.

Keywords:

biometric attributes , speaker verification , Gaussian Mixture Model (GMM) , Support Vector Machine (SVM) , Decision Trees (DT) , Ensemble Classifiers

Authors

Mohammad Mosleh

Young Researchers and Elite club, Dezfoul Branch ,Islamic Azad University, Dezfoul, Iran

Faraz Forootan

Department of Computer Engineering, Dezfoul Branch, Islamic Azad University, Dezfoul, Iran

Najmeh Hosseinpour

Young Researchers and Elite club, Andimeshk Branch, Islamic Azad University, Andimeshk, Iran