Dec. 4, 2008 -- Speech analysis might be flavor of the month on TV, but one expert says popular crime shows may be giving the public an unrealistic idea of what the science can do.
"Most dramas include this kind of thing [and] there's a great deal of poetic license in what they do," said linguistics and phonetics expert Paul Foukles of the University of York.
Foukles, who has provided training in forensic speech analysis for the U.S. Secret Service and the FBI, spoke at a meeting of the ARC Network in Human Communication Science in Sydney.
Forensic speech science, which involves acoustic, phonetic and linguistic analysis of recordings, can be used to try and narrow down possible suspects. It can also be used to determine how well a suspect's voice matches a criminal's voice.
But, said Foukles, a recording of speech can never be used to identify someone with certainty on its own.
Related Content:
"There is no such thing as a voice print," he said. "It's a very very dangerous term. There is no single feature of a voice that is indelible that works like a fingerprint does."
Many different factors influence how people speak at any particular time and place.
"If you're tired or if you have a cold or if you're speaking on a phone against traffic in the background you do all sorts of things to the voice, which make it phonetically very different from time to time," said Foukles, who also works as a freelance consultant for a private forensic speech science laboratory.
"The features of speech and language are such that you can't use them as a marker of identity to identify one person and exclude all other people under normal circumstances. People's voices overlap."
In addition, said Foukles, acoustic analysis using software must contend with the problems associated with the recording. For example, voices are often analyzed from wire taps, covert recordings taken from someone's pocket or messages from a voicemail.
While speech analysis cannot be used to identify someone on its own, it can be used to eliminate suspects, said Foukles.
Phonetic and linguistic analysis can be used to analyze component parts of voice such as pitch, how vowels and consonants are pronounced, voice quality and even vocabulary.
If the forensic investigator is lucky, they will be able to identify certain voice features, pathological features that are difficult to disguise, such as pronouncing 'r's as 'w's, said Foukles.
One issue of concern is the use of software to analyze the acoustic properties in someone's voice to determine their state of mind, including if they are lying.
Foukles said such "lie detectors" or "voice stress analyzers" are based on "questionable principles" but are now being used for security purposes.
"Moscow airport is using this as a way of detecting whether someone is a terrorist threat or not," said Foukles. "I think that is a very worrying development."
Related Links:
our sites
video
mobile
shop
stay connected
corporate
