Абстрактный

SPEECH AND SPEAKER IDENTIFICATION FOR PASSWORD VERIFICATION SYSTEM

Kirti A. Yadav , Minakshee Patil

The voice signal contains lot of information. Direct analysis and synthesize of this speech signal becomes complicated. Therefore voice and speech processing approaches generally have feature extraction and feature matching concept. In computer science, speech recognition (SR) is the translation of spoken words into text. The term voice recognition refers to finding the identity of "who" is speaking, rather than what they are saying. For password verification systems we require both speech as well as speaker identification. This approach is implemented in this paper. Here Mel-frequency cepstral coefficients (MFCC) of recorded speech are stored and then trained accordingly to obtain speech as well as speaker verification for password verification system. With the help of the Euclidean distance we measure the distance between two points of trained data and test data to verify a password spelled by a specific speaker