Expert opinion is increasingly being sought in the legal process as to whether two or more recordings of speech are from the same speaker. Forensic speaker identification can be very effective, contributing to both conviction and elimination of suspects on the basis of their voice. Yet there is still a considerable lack of understanding as to what it involves; what constitutes appropriate methodology; what it can achieve; and what its limitations are. The aim of this book is to address these questions with the appropriate technical precision. This book clarifies the problems of inferring identity from speech under the less than ideal conditions typical in forensics. It will help the reader to appreciate:

the complexities of voice sample comparison

the probabilistic nature of the technique, which should be Bayesian in approach

the difficulties introduced by differential variation within and between voices and the generally poor degree of control over forensic samples

the necessity for both acoustic and auditory comparison, and for the comparison of linguistic and non-linguistic features

the expertise required in linguistics, acoustics and statistics
Drawing on his many tears of teaching experience, the author presents a clear and accessible account, suitable for readers with no prior knowledge of forensic speaker identification or the sub-disciplines - phonetics, linguistics, statistics, speech acoustics and speech perception - which inform it. Examples are given from several languages, including American, British and Australian English, Japanese and Chinese. This is essential reading for the legal profession, law enforcement officers, and phoneticians, speech scientists and linguists, who have an interest in how identity is encoded in speech.

Buy it

Voice-based evidence is an important part of many criminal investigations and has commonly included such things as threats left on an answering machine, a robbery caught on videotape, or a confession recorded during a police interrogation. In the technological age of mobile telephones, voicemail and voice-recognition software applications, the potential for voice-based evidence continues to increase. This book is designed for the experts who must be able to properly handle this type of evidence, decipher what is said, and identify the people who have been recorded. It outlines practical methods for voice identification and covers key areas such as authentication of speech recordings and voice stress analysis.

Buy it