Voice analyst Tom Owen was the first to consider the tapes using a program called Easy Voice Biometrics.
This program listens to a call or voice sample and processes the sounds just like our IVR. The difference is that while an IVR system breaks down and compares the voice to a database in order to determine what the voice is saying, Owen’s program runs a comparison against only one voice (in this case, Zimmerman’s).
You can think of it kind of like translating a book from Russian to English. An IVR system knows the book is written in Russian so it goes through and translates word-for-word using a dictionary. Owen, however, is trying to prove that the book is Russian so his program runs each word through a Russian-English dictionary to see if it’s a match.
Owen isn’t the only expert on the case, though.
Stuart Allen, another forensic-audio examiner, told Discovery News that (with the exception of identical twins) voiceprints are incredibly unique, just like our fingerprints:
“Unless you’ve had surgery, you can’t disguise the characteristics of your vocal cords or mouth structure.”
Unlike IVR, which has only been around for about 30 years, forensic acoustic software has been around since WWII (when we used it to identify enemy voices on radio broadcasts).
Allen, who uses a different program alongside Easy Voice Biometrics, called Trawl, inputs voice samples into his software to create a voiceprint that he can later compare to other existing voiceprints.
So, in simpler terms, Trawl creates a dictionary based on a book and then compares that dictionary to others to make a language match.
…Read the unexpected conclusion in “Voice Tuning”…