Untangling the talkUntangling the talk
All In The Mind
What humans say to each other is perplexing at the best of times. But what if all you have is sight, no sound? Deaf people 'see' language through signing and lip-reading -- with surprising brain scan results. Forensic linguists often only have sound, no sight -- working with grisly sound recordings from murder scenes and covert criminal deals, to sleuth for suspects. Tune in to untangle some tricky talk.
30:06•13 Dec 2008
Untangling Tricky Talk: Forensic Voices, Deaf Brains and the Science of Speech
Episode Overview
- Forensic phonetics can help interpret messy, noisy recordings from crimes and covert operations, but there is no such thing as a perfect voice fingerprint.
- A single voice can change dramatically with context, illness, emotion and environment, so juries need realistic expectations of what experts can say.
- New voice risk analysis tools promoted as lie detectors lack strong evidence and may be no more accurate than chance, raising ethical concerns.
- Skilled deaf speech readers combine sharp visual motion sensitivity, memory and rich language knowledge to make educated guesses from lip and face movements.
- Brain imaging suggests signed and spoken languages are handled in very similar ways by the brain, which challenges arguments for restricting sign use before cochlear implantation.
“Our voices are very, very malleable.”
What can we learn from those who have battled addiction to misunderstanding and miscommunication? This episode of All In The Mind jumps straight into how tricky human speech can be, especially when you only get half the picture. Host Natasha Mitchell chats first with forensic linguist Dr Paul Fawkes from the University of York, who spends his time working with some very dark material.
From mobile phone recordings of attacks to threatening voicemails and covert tapes of drug deals, he explains how forensic phonetics helps police work out who might be speaking and what is actually being said. He warns against the myth of a “voice fingerprint”, noting that, “our voices are very, very malleable”, shifting with illness, emotion and background noise.
You’ll also hear how dialect clues helped narrow down the infamous hoax tapes in the Yorkshire Ripper case, and why new “voice risk analysis” lie-detector software worries him. Then the perspective flips. Emeritus Professor Ruth Campbell from University College London turns to what happens when sound disappears altogether. She outlines how skilled deaf lip readers — or speech readers — can “hear by eye”, picking up tiny movements of lips, tongue, head and body.
Her team has found that good lip readers are especially sensitive to visual motion and rely heavily on memory and strong language skills. She breaks down why some sounds are easy to see (like “Fred” versus “thread”) while others blur together, and how signed languages are now recognised as fully structured languages in their own right.
Brain imaging studies add another twist: watching speech and sign activates many of the same brain areas as hearing speech, challenging old ideas about “hearing centres” and feeding into heated debates on cochlear implants and language access for deaf children. If you’ve ever misheard someone and felt silly, this conversation might make you wonder: how much of what you think you’ve understood is just a clever guess your brain stitched together?
Do you want to link to this podcast?
Get the buttons here!
More From This Show
The latest episodes from the same podcast.