National Public Radio (NPR)
November 13, 2002 Wednesday  

Copyright 2002 National Public Radio (R). All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to National Public Radio. This transcript may not be reproduced in whole or in part without prior written permission. For further information, please contact NPR's Permissions Coordinator at (202) 513-2000.  
National Public Radio (NPR)  

SHOW: All Things Considered (8:00 PM ET) - NPR  

November 13, 2002 Wednesday

LENGTH: 683 words

HEADLINE: Steve Cain discusses the process of forensic tape analysis

ANCHORS: LYNN NEARY

BODY:
LYNN NEARY, host:

Joining us now is Steve Cain. He's the president and CEO of Forensic Tape Analysis Incorporated. He joins us from his office in Lake Geneva, Wisconsin.

Mr. Cain, can you explain to us what methods are used to analyze tape in order to determine if a recording such as the one released yesterday of Osama bin Laden really is Osama bin Laden? Mr. STEVE CAIN (President and CEO, Forensic Tape Analysis): What we normally do is get the best copy, we will listen to it. If we can extract enough words to compare it, we will then try to get a verbatim voice sample. We normally can get this through the courts or a court order, or we will at least try to get the voluntary cooperation of the speaker and have him repeat the same words, so it's basically a verbatim speech comparison, where you look at various words on what they call the voice spectrograph, which is the instrument which measures and displays the vocal patterns of the speaker's voice.

NEARY: What might stand out for you when you're listening?

Mr. CAIN: Most of the speech information that the expert relies on when he's identifying or eliminating someone's voice would involve the vowel sounds, or what they call phonemes. That's what actually gives you the voice prints that you see on a chart or a computer screen. So you're listening for any aspect of the voice that makes it somewhat unique, whether it be mispronunciations of certain words, a higher- or a lower-pitched voice. It's all of these factors that come into play.

NEARY: This is the tape that was released yesterday, a portion of it. Let's play that.

(Soundbite of audiotape)

Mr. OSAMA BIN LADEN: (Foreign language spoken)

NEARY: Now just listening to that, is that a good enough copy that you could perhaps make some kind of determination listening to that?

Mr. CAIN: Until we've pulled the sound off and take a look at it in the voice spectrograph, we won't know whether or not we have sufficient words so that we could do a comparison. It does sound, although it is somewhat--it has static in the background and a lot of various cracks and pops, etc., that should not preclude the expert from still doing a proper voice comparison. And it's a matter of getting a hold of everything that was said two days ago and seeing which ones can be used to do the voice print analysis.

NEARY: Now how much does this just depend on the human ear, apart from getting a voice print analysis? I mean, how much can you tell just by listening? And how often do you listen to these tapes?

Mr. CAIN: No, we listen to the tapes no matter what language is involved in every case. Many times the human ear is pretty accurate. The problem is it can be fooled, and that's why years ago law enforcement contacted Rich Little, the famous mimic--we got him to provide disguised voices of various movie stars, and then we actually recorded the movie stars saying the same words. And in each case, Rich Little's voice, even though to the ear it sounded like it was an authentic voice of maybe Cooper or Humphrey Bogart, in fact under the voice spectrograph, it was obviously Rich Little's own voice.

NEARY: Can you ever be absolutely 100 percent certain that you have successfully determined the authenticity of a voice?

Mr. CAIN: Yes, I have occasionally, both with the Secret Service when I was with the government in Washington, DC, and also in private practice. Occasionally, we have rendered positive opinions either for identification, if we think it is the speaker's voice, or for elimination, if we think that perhaps the wrong person, you know, has been arrested or indicted. So it goes both ways, but in the majority of the cases, the expert still is going to be pretty much required to provide a qualified opinion because there rarely are sufficient words of good enough quality to positively eliminate or identify him.

NEARY: Thank you very much, Mr. Cain.

Mr. CAIN: OK. Thank you.

NEARY: Steve Cain is president and CEO of Forensic Tape Analysis, Inc. He joined us from his office in Lake Geneva, Wisconsin.