Speaker verification using mel-frequency cepstrum coefficient and linear prediction coding
Shubhangi Agrawal, Devyani Mishra · 2017 International Conference on Recent Innovations in Signal processing and Embedded Systems (RISE) · 2017
Identity verification is one of the most important concern in today's era of information technology. Conventionally, it is done using keys or password which could be stolen and are not suitable for critical purposes. To overcome these problem various biometric verification system have emerged, such as iris recognition, figure print recognition and speaker recognition system. In this paper the author has provided brief study of feature extraction method, linear prediction coding (LPC) and Mel-Frequency Cepstrum Coefficient (MFCC) in speaker recognition system. It also presents implementation of various blocks of MFCC using the following design suit : 1. Design is done using VHDL and simulated using ISE 2. Device technology used 90nm. 3. Device family SPARTAN3E; Device XC3S500E and Device package FG320.