Voice Similarity and the Effect of the Telephone

a study of the implications for earwitness evidence

Voice similarity and the effect of the telephone:
a study of the implications for earwitness evidence

Department of Linguistics, University of Cambridge
Award no. RES-000-22-2582
January-December 2008
Principal investigator:
Francis Nolan
Kirsty McDougall
Research Assistant:
Toby Hudson

In certain crimes, a perpetrator may be heard but not seen. For example, the perpetrator may be masked, or the victim blindfolded, or it might have been too dark for a witness to see clearly, yet the voice of the criminal was heard. In such cases, 'earwitness' evidence may be provided by a 'voice parade'. The witness is asked whether he or she can pick out the voice heard during the crime from a line-up of recordings which includes the suspect's voice and a number of foil voices. Other crimes for which earwitness evidence might be possible are those where the perpetrator is heard over the telephone, for example threatening phonecalls, ransom demands, stalking, blackmail and extortion. It is not yet clear whether a voice parade can be usefully employed in such cases as there is little research into listeners' ability to recognise voices heard over the telephone. This will be addressed in this project through two studies, the first investigating the extent to which telephone transmission affects a listener's ability to distinguish among similar-sounding voices, and the second examining the reliability of a listener's performance in a voice parade for ordinary versus telephone exposure to a voice. The project will also investigate the acoustic correlates of voice similarity to lay the foundations for a framework for the description of voice similarity applicable to voice parades.