Quote:


As with polygraphy, there is no peer-reviewed research supporting the validity of "Layered Voice Analysis" for lie detection, love detection, or any other purpose for any age group.

Would normal pitch or patterns of a person's voice be construed as stress?  take an adolescent whose voice pattern can change constantly because of growth and hormonal changes.  so where is the considered baseline on where the voice is considered normal and considered stressed?  sounds like a guessing game to me.

Quote:


An audio recording of Mr. Maschke's voice is available on the home page of this website.  Go ahead and give it a try.  Mr. Maschke may very well oblige you, but why should he.  As you note it isn't a full blown scientific test.  The only science to it is that it records the unexplored world of vocal stress patterns.

Quote:


I would venture that these "amazing results" are the fruits of deception.  It is not that the "test" works.  It is that the criminals providing a confession believe it works.

If you are interested, I have for sale, beach-front property in Arizona.

If George gives them his audio recording and they do the LVA test, it would show that he is not affraid to find out the truth, even if it isn't a full blown scientific test.  I for one would like to see the results.  I have been hearing a lot of second hand stories about the amazing results some of my police friends have been getting using the LVA software.

Quote:


But such a small study would be statistically meaningless and would reveal nothing about the validity of Layered Voice Analysis.
   
Quote:


If you review this message thread, you'll find that I have, in fact, investigated Layered Voice Analysis. And I found no credible evidence to support the claims of those marketing it. If you are aware of relevant information that you think I have ignored, please direct me to it.

Quote:


But the experiment you suggested would reveal nothing about the validity of Layered Voice Analysis, not only because of the small sample size, but also because of the virtual impossibility of independently verifying ground truth in a screening scenario such as you proposed: the questions asked in pre-employment screening polygraph examinations typically involve using and selling drugs, espionage, and sabotage. But how can one prove that one has never done these things? You cannot prove a negative.

A more revealing experiment that I might be willing to partake in is this: on videotape, I will flip a coin 100 times, with the result being shown on camera. After each flip, I will make two statements: 1) The coin is heads up; 2) The coin is tails up. These statements will also be audio recorded.

The purveyors of Layered Voice Analysis will receive the audio recording, perform an analysis of my voice, and determine the result of each coin toss. The results will then be placed on-line for the edification of all.


Quote:


It's not concern about the possible results of the experiment that you suggested which leads me to conclude that it would not be worthwhile, but rather the fact that the results would shed no light on the validity of Layered Voice Analysis.

By contrast, the experiment I have suggested would produce results that could readily be compared against ground truth and from which some logical inference(s) might be possible.

Quote:


And since George has neither cajones nor integrity, you can rest assured he will not be taking any more tests.

To Neutral Observer

I hadn't thought of it that way.  I actually meant vindicate George by results agreeing with his claim that he was telling the truth.  To be believed by something you don't believe in would be a bit awkward to say the least.
Thanks for pointing that out, it would take someone with considerable cajones and integrity to have taken up my reasonable suggestion in these circumstances

Quote:


Right, that would make George a three time loser.  And it would strain anyone's credulity to think he was anything but the liar he has been diagnosed as being on both previous polygraph tests.

Well it would give you the opportunity to compare the responses to your answers like for like.
   
It would also protect you from a charge of being guilty of "contempt prior to investigation" which is your current position from an epistemiological perspective. 

I don't know whether this stuff works or not so to hear a first hand report of a direct comparison between these two systems would have been helpful.

But I do understand the difficulty from your own personal stance.  Should this equipment actually vindicate your polygraph experience where would you go from there? 

What purpose would be served by such an undertaking? I think it would be about as interesting as posing the questions to a Magic 8-Ball.

Instead, those claiming that Layered Voice Analysis is a reliable test for deception should provide proof, something they have thus far utterly failed to do.

Hey George,

Why don't you undergo exactly the same questioning with LVA that you had with the polygragh?
I think it would be really interesting to hear your opinions on the whole experience from a firsthand perspective.
Would that be possible?

Okay, if this technology is as good as the creator claims, then can an insurance company deny a claim based on the results of the LAV?  How do the results effect law enforcement?   ???

What a bloody Joke!!!!!!!!

The Software in question was originally marketed in Israel as a child's TOY!! (Ex-Sense Pro) and sold in toystores and bookstores without that much success.

Please keep in mind that here in Israel, we also use handwriting analysis, amulets, and magic spells (blessings) as a matter of routine within our government agencies.

The Hebrew word for "sucker" is "freir", and only freirs would be taken in by such deceptive marketing!

See Colin R. Johnson's EE Times article, "Lie detector glasses offer peek at future of security" for recent news on "Layered Voice Analysis":

http://www.eetimes.com/story/OEG20040116S0050

A lively discussion of this article is to be found on Slashdot.org here:

http://slashdot.org/articles/04/01/20/1857249.shtml

Quote:


Rick,

Thanks. Unfortunately, having only worked in the private sector, I'm ill equiped to search this. It's likely some others here could assist though. Being an EE by profession, I am most interested in how this was approached. Especially given the limited technology back then. These days the most sophisticated DSP is dirt cheap.

-Marty

Marty wants to know where study was published. The only publication I am aware of was a Technical Report to the US Army's Limited Warfare Laboratories in 1973. I used to have a copy and if I find it, can provide the document number, but if you're handy retrieving federal documents from their archive services it should be locatable. If its interesting, there were several EE's associated with this sudy, as well as Psychology PHD's (and one ABD). And, of course, appropriately qualified personnel from several selected federal agencies who had an interest.

Which brings up an interesting point: anyone who has a background in math/science/engineering is going to be skeptical of unsupported claims about the efficacy of the polygraph. My guess is that the "Dr." who made a post is a Ph.D. - in a non-science related area, like psychology, education, and so on. The EdD's, "doctor" of education, is the lowest, I mean the absolute lowest, form of any doctorate, and they tend to overuse the "Dr." title. But I digress.

Any "Dr." who makes claims about how accurate or reliable some instrument is surely knows those claims are going to be challenged, and that is part of the scientific method. Claims without supporting evidence are BS, and that is a fact.

Quote:


Double blind eh?  Where is the study published? I would be most interested in understanding it (especially being an EE).

-Marty

I worked in the polygraph/voice analysis field in the early 70's and performed double blind analyses of tape recordings of interrogations in both English and in a non-Romance language to ascertain truth or deceit with a prototype voice stress analyzer. The research work was paid for, in part, by several federal agencies. The analyses that I performed resulted in an assessment of 83% accuracy, based on post-interrogation determination of the truthfulness of taped responses.

I felt good about the 83%, it was comparable to polygraph and a whole lot more efficient, but it wasn't good enough to call this technology "lie detection", which is why we continued to call it voice stress analysis.

I read the LVA brochure - very interesting and a quantum leap in the technology. I'd love to know the specifics of their "18 voice parameters", since I was pretty well versed in this stuff once. I find their claims to be a great stretch and am very curious how they achive this. They don't have to worry about patent infringement, I have my own and am not interested in theirs.

Saidme wrote on Aug 1^st, 2003 at 7:03pm:


Just curious..what would make you think I was actually close to giving you a 'hint of a compliment'?   

If laughing at how even a polygraph apologist such as yourself turns his back on his own kind is some sort of compliment, feel free to pat yourself on the back alittle more.   

saidme,

How many times over how many years= what percent of accuracy?

Yes I do!  I've had several exams over the years where no confession was obtained, yet additional evidence was obtained later which verified the polygraph results.  You can say whatever you want but years of success can't be just ignored or thrown away.

Saidme,

Does it really matter what kind of machine or system is used?

It's the confession that the proof, right?

Do you really think that the polygraph is any more or less accurate?

I've never used LVA and wouldn't know what one looked like even if it were hooked to George.  As I stated earlier in this thread, I put little credibility in CVSA or LVA.  I'm sure the next statement from one of you will ask how I can make such an assessment without ever having used it.  Because we're in America where I can voice my opinion on anything.  I assume LVA training is probably quite similar to CVSA training.  Which means, within a very short period of time you can have your secretary out running CVSA and LVA exams.  Sounds dangerous.   

Saidme,

Do you think that LVA can distinguish truth from deception, based solely on it technology (meaning no post interrogation)?