Design of Post-Mapping Fusion Classifiers for Voice-Based Access Control System

This paper introduced voice-based biometric system for access control. The ability to verify the identity of a person by analyzing his/her speech, or speaker verification, is an attractive and relatively unobtrusive means of providing security for admission into an important or secured place. In the field of speaker verification, the main objective is to achieve the highest possible classification accuracy. The proposed system focused on combining the classification scores. Features are extracted from raw data and can be diverse. Therefore, in post-mapping fusion, each feature set is modeled separately, and the output score of the classifiers are combined to give the overall match score. Furthermore, for each classifier score, an a priori weight is set based on the level of confidence of the feature set and the classifier. Three different feature extractions involved in this work are Liner Prediction Cepstral Coefficients (LPCCs), Mel Frequency Cepstral Coefficients (MFCCs) and Perceptual Linear Prediction (PLP) coefficients. While the classifier used in this study is Support Vector Machines (SVMs). Experimental result confirms that in terms of false acceptance rate (FAR) and false rejection rate (FRR), the Post-Mapping Fusion Classifiers is effective to use in the proposed system.