Voice Analysis on Mobile Phone’s Voice Recorder Using Audio Forensic Method

Abstract

This research provides an example scenario of applying digital forensic techniques for voice pain 
recognition. Greetings compare the sound in the evidence and the sound suspect on mobile phone media with a voice case study woman. In this study, PRAAT application was used to help with the audio comparison process  from known samples and unknown samples. The results showed that there are no three F0's in the low,  medium, and high tones of the word “proyek” from the same comparison speaker as the original speaker.  Based on the F1-F4 analysis of the voice data with PRAAT, the difference between the original speaker's voice  and the voice of the comparison speaker can be seen. No similarity was found between F1-F4 in the low,  medium, and high-pitched voices of native speakers and comparison speakers. From these results, it can be  seen that there are no F1-F4 vocal formants in the word “terlibat” from the same comparison speaker as the  native speaker so that the difference becomes apparent.