Using Logarithmic Spectrographs of Vowels as a Key for Forensic Speaker Recognition

Namvar Fargi, Mojtaba; Sharifi, Shahla; Sharifi, Shahla; Estaji, Azam; Pahlavan Nezhad, Mohammad Reza; Estaji, Azam; Meshkatoddini, Mehi

Using Logarithmic Spectrographs of Vowels as a Key for Forensic Speaker Recognition

Authors

Mojtaba Namvar Fargi ¹

Shahla Sharifi ²

Azam Estaji ²

Mohammad Reza Pahlavan Nezhad ²

Azam Estaji ²

Mehi Meshkatoddini ³

¹ Ph.D., Department of Linguistics, Ferdowsi University of Mashhad, Mashhad, Iran

² Associate Professor, Department of Linguistics, Ferdowsi University of Mashhad, Mashhad, Iran

³ Professor, Department of Linguistics, Ferdowsi University of Mashhad, Mashhad, Iran

Abstract

Forensic phonetics is a subfield of forensic linguistics in which acoustic information and phonetic features of phones are investigated for completing the forensic cases where one of the existing evidences is related to the guilty. One of the most important tasks of a forensic phonetician is forensic speaker recognition. For doing this, the phonetician is asked to estimate the degree of similarity between the given records of the guilty’s and the suspect’s speech, and determine whether these two sound evidences match to each other or not. The objective of this study, which was conducted on the sound data from 10 Persian native speakers of both sexes, was to investigate the possibility of using Logarithmic spectrographs of vowels as a key for forensic speaker recognition tasks. The results showed that using these logarithmic spectrographs may be a useful means with perfect reliability in the tasks related to forensic speaker recognition.

Keywords

Forensic linguistics

Forensic phonetics

Speaker recognition

Acoustic phonetics