Voice Based Database Management System Using DSP Processor

Abstract

Security is provided to customers through PIN/ID protection, to secure their data and information using password. This method require user to authenticate them by entering password. There are cases of fraud and theft when people can easily know the password. But there is a existing technology known as Bio-metric Identification System. It uses an individual's bio-metric characteristics, that is unique and therefore can be used to authenticate the user authority access. This invention is an implementation of speaker verification for attendance monitoring system using DSP Processor. First, speech signal will go to pre-processing phase, where it will remove the background noise then, speech signal's features will be extracted using Mel Frequency Cepstral Coefficients (MFCC) method. Then, using Hamming Window, the features will be matched with the reference speech in database. The speaker is identified by comparing the speech signal from the tested speaker. The main focus of this invention is speaker verification, which is compared between speech signal from unknown speaker to a database of known speaker using utterances. The speaker identification is used in this invention for creating database and identifying the students for maintaining attendance record. LCD Display is interfaced with Processor to show the presence of the students for a particular subject. So this can be used for monitoring the attendance of students. Also defaulter student’s list is find out according to the criteria, and it is maintained in MS Excel sheet. Future scope for this can be, informing the monthly attendance to their parents through a text message on their mobile phones using a GSM Module interfaced to processor.
DOI: 10.17762/ijritcc2321-8169.1505114