Analysis of Speaker Identification Systems

International Journal of Artificial Intelligence and Knowledge Discovery

Mandeep Singh Walia¹ 

关键词: biometrics;    feature extraction;    dynamic time warping;    Gaussian mixture modelling;   

DOI  :  

学科分类：建筑学

来源: RG Education Society

【 摘 要 】

Normal0falsefalsefalse EN-USX-NONEX-NONESpeaker Identifications are probably the most natural and economical methods for solving the problems of unauthorized use of computer and communications systems and multilevel access control. With the ubiquitous telephone network and microphones bundled with computers, the cost of speaker recognition might only be for software.A recognition System is a pattern recognition system that operates by acquiring data from an individual and then does pre-processing, segmentation, extracting the feature set and comparing the feature set against the template set in the database.Speaker identification is a performance biometric, i.e., performs a task to be recognized.  Voice, like other biometrics, cannot be forgotten or misplaced, unlike knowledge-based (e.g., password) or possession-based (e.g., key) access control methods. Old method of identification of speaker was identification through spectrograms. But this method had lot of problems and difficulties. To overcome the shortcomings of previous technique, the idea of automatic speaker identification is suggested. In this features are first extracted from speech samples and then after modeling, the samples are stored in a speaker database. For speaker identification, when matching required, the features are again extracted from speech samples and are compare with stored database. A decision is taken on the basis of this match to accept or reject the sample.In this paper, I propose speaker identification systems based on Dynamic Time Warping (DTW) and Gaussian Mixture Modelling (GMM).There are a number of techniques for feature extraction and speaker modeling and all have their own advantages and disadvantages. In the present work, feature extraction technique and two speaker modeling methods are discussed. Also an attempt has been made to compare these techniques./* Style Definitions */ table.MsoNormalTable{mso-style-name:"Table Normal";mso-tstyle-rowband-size:0;mso-tstyle-colband-size:0;mso-style-noshow:yes;mso-style-priority:99;mso-style-qformat:yes;mso-style-parent:"";mso-padding-alt:0in 5.4pt 0in 5.4pt;mso-para-margin:0in;mso-para-margin-bottom:.0001pt;mso-pagination:widow-orphan;font-size:11.0pt;font-family:"Calibri","sans-serif";mso-ascii-font-family:Calibri;mso-ascii-theme-font:minor-latin;mso-fareast-font-family:"Times New Roman";mso-fareast-theme-font:minor-fareast;mso-hansi-font-family:Calibri;mso-hansi-theme-font:minor-latin;mso-bidi-font-family:"Times New Roman";mso-bidi-theme-font:minor-bidi;}

【 授权许可】

Unknown