Application of Digital Forensics to Identify Human Voices Using the System Development Life Cycle (SDLC) Method

• Published in: International Journal of Education and Management Engineering Vol. 12; no. 1; pp. 29 - 37
• Main Authors: Misriani, Nurtanio, Ingrid, Wahid, Omar
• Format: Journal Article
• Language: English
• Published: Hong Kong Modern Education and Computer Science Press 08.02.2022
• Subjects: Evidence; Forensic computing; Forensic sciences; Persuasive Discourse; Reading; Similarity; Sound; Systems development
• ISSN: 2305-3623; 2305-8463
• DOI: 10.5815/ijeme.2022.01.04

This study aims to identify digital audio forensics using the System Development Life Cycle (SLDC) method which is used as a reference in the audio forensic investigation process. The process of testing the application of the framework that was carried out succeeded in identifying audio evidence with the identification results that subject x sampling (known) was identical to recorded evidence (unknown) with the results obtained for more than 4 identical words and supporting the prosecution hypothesis. Also, the results of the feasibility test of a framework that has been developed as a reference standard for comparison of frameworks related to other audio forensics, show that the framework that has been developed has a more complete stage to be used in the audio forensic investigation process. The results of Spectrogram analysis and Pict analysis on values matrix cross similarity level of evidence with audio subject Nasri4Y has the highest similarity value 0.9575822. The results of reading the audio evidence matrix with audio subject Bakrim5Y have the lowest similarity value 0.48924464. The results of reading the matrix, audio subject- B with audio subject Bakrim3Y have the highest similarity value of 0.9287775 because it is a sample voice from the same person. The results of the reading of the matrix, Nasri2Y audio subject, and Nasri4Y audio subject have a similarity value of 0.9575822 because they are sound samples from the same person. The results of reading the matrix audio subject Nasri2Y with audio subject Nasri4Y have the highest similarity value of 0.9575822, from this result it can be said a significant value because the audio subject Nasri2Y and Nasri4Y have the most similar level of sound samples from other subjects because Nasri2Y and Nasri4Y are sound samples from the same person.