Development Of Speech Recognition System For Forensics Application

Yap, Beng Heng (2015) Development Of Speech Recognition System For Forensics Application. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text (24 pages) Development Of Speech Recognition System For Forensics Application 24 Pages.pdf - Submitted Version Download (583kB)

Abstract

Speech recognition system is a system that able to recognize the speech of the speaker. While speaker identification is able to determine the identity of the person speak from the unknown voice. One of the demanding areas of recognition system is in forensics application usually in the case where a crime has been committed and the voice of the criminal needs to be verified from a recorded message. However the sample of voice signal recorded may contain error such as noise which reduces the reliability of this method. The noise occurs in the environment and the status of the person spoke such as illness and drug infection would directly affect the result of the recognition. With the advance development of digital signal processing, more reliable method are found to processed the voice signal which can increase the accuracy of voice matching. In this project, systems which able to determine the identity of the speakers from an unknown voice are develop using MATLAB. A graphical user interface developed so that user is able to interact with the system in a convenient way. The voice feature of the input speech would be extracted by Mel Frequency Cepstral Coefficients (MFCC) compare with the reference model that registers in database. As a result, matching database voice id will be display in the system. From the developed Speech recognition system with GUI, this system should be able to analyze the input voice signal and fairly recognize voice identity of the speaker.

Actions (login required)

View Item