Speaker Recognition and Performance Comparison based on Machine Learning

Main Article Content

Rajeev Ranjan


The speaker recognition is one of the most emerging field in the area of digital signal
processing. In this paper, analyzed the performance comparison of voice recognition using Mel frequency
Cepstrum coefficient (MFCC) with two methods based on vector quantization (VQ) techniques by Linde
buzo gray (LBG) algorithm and improved weighted VQ method. In the first phase, extract the minor
amount of data points from the voice signal that is used subsequent to represent every speaker is known as
MFCC and in the second phase, for feature matching there are two approaches which have been proposed
here based on VQ techniques for recognition purpose and also comparison of both algorithms are being
done with different time length speech samples to get better recognition rate and improving in efficiency of
the system. The VQ is used as feature matching with traditional LBG Algorithm and improved weighted
VQ algorithm for better recognition. The weight of vector in VQ algorithm to get the weighted distortion
and after compare between traditional and improved weighted VQ algorithm. When the test time ‘1s’ for
traditional VQ algorithm 81.25% whereas improved weighted VQ 87.5 % and if test time ‘2s’ for
traditional VQ algorithm 87 % whereas, improved weighted VQ 93.7 %. That is improved weighted VQ
algorithm gives the better speaker recognition than VQ with LBG algorithm.

Article Details