Forensic phonetics, a form of forensic speech and audio analysis is a relatively new discipline that was founded in the 1970s with the aim of identifying suspects in crime by analysing the characteristics of the human voice and how the sounds of language are produced. It is essentially a field of applied science that combines the effective collaboration of linguists, phoneticians and computer technicians for an effective and thorough analysis of the human voice.

It allows the possibilities of voice identification with the help of computer identification that encompasses auditive forensic analysis, analysis by computer and acoustic forensic analysis or sounds that are distinctive (Eg a physical oral characteristic of an individual such as a tooth loss.)

There are several parameters that would influence voice analysis that are relevant to forensic phonetics. An individual’s emotions, physiology, speaking habits, age, dialect, health and utterance would influence the speech produced while the signal transmission of the speech by electronic means would take into account the environmental influences (such as echo), encoding and signal compression, the acoustic signal, and noises made if an individual is ill (if one has flu or a sore throat).

Forensic phonetics would also analyse and take into account the various parameters that would have a bearing on forensic phonetics. This includes auditive speaker identification, age, instrumental procedures, the parameters of voice identification, the parameters of language, and manner of speaking. In the parameters of language, factors such as dialect, foreign accent, individual characteristics, level of education and socio-economic background must be considered. Factors under manner of speaking include rhythm of speech, speaking rate, fundamental frequency patterns and breathing behaviour.

Pétursson also spoke on the influence of alcohol on the speech and how it has a paralysing effect on normal speech pattern. Where age is concerned, it is difficult to distinguish disguised voices of males within the ages of 20-40. This is due to the individual’s health and whether one is a smoker because smokers tend to sound more mature by 6-7 years. In addition, the voice of males is harder to distinguish than females who have a more regular voice than their male counterparts. Also, disguised voices of males are harder to decipher due to the linguistic competence and sound variances of males that exceed the female voice.

Because criminals in general tend to disguise their voice when speaking, the German police in collaboration with Pétursson have established a database of dialects to aid their efforts in combating crime that involves criminals who disguise their voices. There is growing interest in the applied science of forensic phonetics in crime solving and its importance is vital in forensic speech and audio analysis. Perhaps it is timely that our police consider setting up a forensic phonetics unit.

Tneh teaches English language and literature at the Dept of Modern Languages, Faculty of Creative Industries, UTAR. Comment: letters@thesundaily.com