AN ANALYSIS OF SMARTPHONE’S VOICE RECORDING USING PRAAT (CASE STUDY: FORENSIC AUDIO)

Abstract

Nowadays, technological developments have contributed to the sustainability of human life. Especially in the field of law or forensics which continues to evolve with the times. Sound recording evidence can show the identity of the person whose voice is recorded on the evidence by means of perform audio forensic examination for voice recognition. Therefore, this study is aimed to analyze the sounds by using the comparative method. Technically, this method is proposed to compare the sound of goods evidence with recorded sound for comparison. In this study, the analysis was limited to identify the value of voice recognition on the statistics of pitch and formant. If the results of voice recognition indicates that the voice of the conversation that is inside the evidence is identical to the voice of the perpetrator, then it can be concluded that the voice of the conversation inside the recorded evidence is the voice belonging to the perpetrator so that the voice record can be used as strong evidence in court. As conclusion, the difference between the voice of the original speaker and the voice of the comparison speaker are clearly shown in some parts of pitch and formant. On the other hand, there is no similarity of each word in value of pitch (F0) in the low, medium, and high voices of native speakers with non-native speakers. Thus, the indicators of the validity of the original sound proof (native speaker) look vague and can't even be proven