A new direct access framework for speaker identification system

摘要

We present in this paper a new Direct Access Framework (DAF) for speaker identification system, to identify a speaker based on original characteristics of the human voice. Direct access method is a process to identify an object based on parts of the object itself, the parts called original characteristics. The proposed framework consists of two parts, the enrolment process and the identification process. Phases are as the following: speech preprocessing, speaker feature extraction, feature normalization, feature selection, speaker modeling, direct access method and speaker matching. In this paper, we used Indonesian speaker dataset containing 2,140 speech files, 142 speakers, 97 male and 45 female. The identification accuracy level based on MFCC features is 94.38% and the accuracy of speaker gender-based classification up to 100% based on pitch, flatness, brightness, and roll off features. The proposed framework helped the researcher in speaker identification system domain for implementing their proposed algorithms or model to obtain the best speaker identification system for various dataset. DAF is also could be used as a basic framework for the other multimedia data as well as image or video.