Improved Speaker Recognition for Degraded Human Voice using Modified-MFCC and LPC with CNN

Economical speaker recognition solution from degraded human voice signal is still a challenge. This article covering results of an experiment which targets to improve feature extraction method for effective identification audio with the help data science. Every speaker’s has identical characteristics. Human ears can easily identify these different characteristics and classify audio. Mel-Frequency Cepstral Coefficient (MFCC) supports get same intelligence in machine also. MFCC extensively used extraction. In our we have effectively Linear Predictive Coding (LPC) better accuracy. first outlines frames then finds cepstral coefficient each frame. use convert it numerical value features, recognize efficiently by Artificial Intelligence (AI) based system. covers how features be extracted signal. observed improved Equal Error Rate (EER) True Match (TMR) due high sampling rate low frequency range mel-scale triangular filter. also pre-emphasis effects on when background noise comes