Nemesysco Founder Amir Liberman Is a Charlatan

Nemesysco founder Amir Liberman
Nemesysco founder Amir Liberman

Amir Liberman, the founder of Nemesysco, an Israeli company that internationally markets voice based lie detectors that simply don’t work, successfully pressured an academic journal into withdrawing the Internet availability of a peer-reviewed article that exposes Liberman’s lie detection “technology” for the pseudoscientific flapdoodle that it is.

The Article

Swedish linguists Anders Eriksson and Francisco Lacerda co-authored an article titled “Charlatanry in Speech Science: A Problem to Be Taken Seriously” that was published in the International Journal of Speech, Language and the Law (vol. 14, no. 2 [2007]). Eriksson & Lacerda review several voice-based lie detectors, including Nemesysco’s “Layered Voice Analysis” (LVA) which the U.S. military’s Special Operations Command has purchased and the Los Angeles County Sheriff’s Department is considering adopting. A variant of LVA customized for security checkpoints has reportedly been trialled at Moscow’s Domodedovo International Airport.

Anders Eriksson and Francisco Lacerda
Anders Eriksson and Francisco Lacerda

Eriksson & Lacerda point out that so-called “thorns” and “plateaus” — characteristics of digitized voice recordings that Nemesysco claims reflect emotional states — are merely artifacts produced by the digitization process! With regard to the LVA software, Eriksson & Lacerda note:

Contrary to the claims of sophistication — ‘The LVA software claims to be based on 8,000 mathematical algorithms applied to 129 voice frequencies’ (Damphousse et al. 2007: 15) — the LVA is a very simple program written in Visual Basic. The entire program code, published in the patent documents (Liberman 2003) comprises no more than 500 lines of code. It has to be said, though, that in order for it not to be possible to copy and run the program as is, some technical details like variable declarations are omitted, but the complete program is unlikely to comprise more than 800 or so lines. With respect to its alleged mathematical sophistication, there is really nothing in the program that requires any mathematical insights beyond very basic secondary school mathematics. To be sure, recursive filters and neural networks are also based on elementary mathematical operations but the crucial difference is that these operations are used in theoretically coherent systems, in contrast to the seemingly ad hoc implementation of LVA.

Continue reading Nemesysco Founder Amir Liberman Is a Charlatan

Colorado Television News Program Investigates Computer Voice Stress Analysis (CVSA)

Colorado television station KUSA 9News investigative reporter Jace Larson examines the use of the “Computer Voice Stress Analyzer” (CVSA) in the state:

KUSA – A device used by Colorado law enforcement agencies to identify when someone is lying, may not work and may be costing taxpayers money.

Computer Voice Stress Analyzers (CVSAs) claim to measure changes in a person’s voice that indicate a lie.

However, three recent studies say the device does not accurately tell the difference between a person lying and a person telling the truth.

CVSAs have been used by 21 law enforcement agencies in Colorado.

Studies by the National Academy of Sciences, the International Association of Chiefs of

Police and the Department of Defense question the validity of CVSAs.

In 2006 a University of Florida study found CVSAs, “performed at chance-level for deception, truth and stress.” The same study went on to say, “false positive rates were high.”

Westminster Police Investigator Wayne Read doesn’t agree though. He and members of his department have used the device and swear by it.

“I know how to operate the instrument. I know how the instrument works. I don’t think I could deceive the instrument,” said Read.

Read’s belief in CVSA despite the scientific evidence against it is reminiscent of the dogged belief of polygraph operators in their own pseudoscientific form of lie detection, despite broad consensus among scientists that it has no scientific basis.

Instructors who teach law enforcement agents how to read the test agree.

“It’s not audible to the human ear,” said Ben Conrique.

Conrique works for The National Institute for Truth Verification, the company selling CVSAs.

“Voice Stress indicated whether or not a person is telling the truth,” said Conrique.

And yet the National Institute for Truth Verification admitted before a federal court that CVSA “is not capable of lie detection.”

The National Institute for Truth Verification sells each device for about $10,000.
Agencies in Colorado have spent more than $331,000 on training and equipment.

“You only spend that type of money on something that has a proven success rate,” said Conrique.

However, experts who oppose CVSAs believe the devices do not work and that they lead to false confessions by suspects.

Deputies in Maricopa County, Ariz. suspected Robert Louis Armstrong of triple murder.

They questioned him for 10 hours in 2003. After deputies told him his test showed he was lying, Armstrong confessed.

Evidence emerged later proving Armstrong was out of the state at the time of the murders and he was freed.

He sued the sheriff’s office.

9Wants to Know tried to speak with deputies in Maricopa County but they declined.

Several agencies around the county have decided to stop using the CVSAs and now rely on other methods such as a polygraph test.

Unfortunately, polygraph tests, too, are completely unreliable as a means of lie detection. They are inherently biased against the truthful, yet easily manipulated through the use of simple countermeasures. See The Lie Behind the Lie Detector for a thorough debunking.

“If you think the CVSA is going to tell you whether witnesses or suspects are telling the truth, you’re gravely mistaken,” said CVSA opponent Richard Leo.

Leo is a criminologist and professor of law at the University of San Francisco.

Leo told 9NEWS if a law enforcement agency buys this device, “You’re wasting your money and you’re wasting public money.”

“You might as well be flipping coins or reading tea leaves or reading an Ouija board,” he continued.

The following are agencies that have confirmed to 9Wants to Know that they own or have used CVSAs:

*Boulder Police Department
*Brighton Police Department
*Broomfield Police Department
*Colorado Division of Wildlife
*Douglas County Sheriff’s Office
*El Paso County Sheriff’s Office
*Englewood Police Department
*Federal Heights Police Department
*Fort Morgan Police Department
*Glenwood Springs Police Department
*Golden Police Department
*Grand Junction Police Department
*Lakewood Police Department
*Lamar Police Department
*Longmont Police Department
*Moffat County Sheriff’s Office
*Northglenn Police Department
*Sterling Police Department
*Thornton Police Department
*Westminster Police Department
*Yuma County Sheriff’s Office

For further reading and video links, see the KUSA 9News website’s feature page, The Truth About Lies. and for discussion, see the CVSA and other Voice Stress Analysis Applications forum of the message board.

Oklahoma Study Finds Voice Stress Analysis “Testing” No Better Than Random Chance

The National Institute of Corrections has published a research study on voice stress analysis (VSA) conducted by the Oklahoma Department of Mental Health and Substance Abuse Services with support from the National Institute of Justice. From the abstract (emphasis added):

…The goal of this study was to test the validity and reliability of two popular VSA programs (LVA [Layered Voice Analysis] and CVSA [Computer Voice Stress Analyzer]) in a “real world” setting. Questions about recent drug use were asked of a random sample of arrestees in a county jail. Their responses and the VSA output were compared to a subsequent urinalysis to determine if the VSA programs could detect deception. Both VSA programs show poor validity – neither program efficiently determined who was being deceptive about recent drug use. The programs were not able to detect deception at a rate any better than chance….

The 128-page report, “Assessing the Validity of Voice Stress Analysis Tools in a Jail Setting” by Kelly R. Damphousse, Laura Pointon, Deidre Upchurch, and Rebecca K. Moore, dated 31 March 2007, may be downloaded from here (2.1 mb PDF).

Daytona Beach News-Journal Discredits Voice Stress Analysis

In a 28 April 2007 editorial cryptically titled, “Wired Policing Stresses Voices, More,” the Daytona Beach News-Journal skewers local law enforcement agencies’ reliance on voice stress analyzers:

As the urban legend goes, police interrogating a suspect put a colander on the suspect’s head, run wires from the colander to a copy machine, make a meaningless copy, look at it, then tell the suspect he’s lying. The suspect, of course, doesn’t know the whole thing is a hoax. Frightened by the amazing lie detector, he confesses.

The legend might as well be true. Local police agencies, including the Volusia County Sheriff’s Office and the Daytona Beach Police Department, use so-called “voice-stress analysis” as part of their interrogation techniques. The Flagler County Sheriff’s Office plans to use it as part of its job-interview process. A study by the Defense Department has concluded that such analysis is bunk — no more reliable than the colander trick. Nevertheless, police agencies spend thousands of dollars on the equipment and justify it as a legitimate part of police work. The Daytona Beach Police Department just spent $32,000 for the equipment.

It’s the best, the most reliable voice stress-analyzer out there,” said Lt. Gorgi Colon, who, with Sgt. Paul Barnett, was tasked by Daytona Beach Police Chief Mike Chitwood to research the technology. Gleaning the studies wasn’t part of the research. Calls were placed instead to the Volusia sheriff’s office and to the West Palm Beach company that manufactures the “analyzers” and dominates the field by aggressively marketing itself as the “world leader in voice-stress analysis.” As part of its marketing, the company trains individuals in agencies that buy the product.

The company’s name — the National Institute for Truth Verification — makes it sound like an academy. It isn’t. It’s a privately held company directed by Bill Endler, a retired Indiana police chief who spent four months interrogating suspects at Abu Ghraib prison in Iraq and in Baghdad’s Green Zone between October 2003 and January 2004. His contract there was not renewed, and the Pentagon, based on its own Department of Defense Polygraph Institute studies, distrusts the voice-stress technology. “In our evaluation,” Mitchell S. Sommers, an associate professor of psychology in Arts & Sciences at Washington University in St. Louis, told a university publication in 2004, “voice-stress analysis detected some instances of deception, but its ability to do so was consistently less than chance — you could have gotten better results by flipping a coin.” Sommers’ research was paid for by the Defense Department.

“What it boils down to,” Endler said, “is nothing more than a turf war. We are taking business away from them.” But the Polygraph Institute isn’t a business. It merely supervises government agencies’ use of polygraph technology, and hasn’t “accepted” voice analysis (in Endler’s word) as legitimate. It could well be a turf war. It could also be that the technology is the gimmick that the Defense Department says it is: Aside from Sommers’ study — peer-reviewed and published in an academic journal in 2006 — the research is scant. Endler also criticized the Defense Department for relying on laboratory experiments rather than actual cases of interrogation. But the Sommers study published in 2006 included “field questioning.” The results were not statistically different from scripted questioning.

Still, local agencies have no qualms about using the technology, which is inadmissible in court (whether it’s traditional lie detectors or voice-stress analysis). Volusia’s and Flagler’s sheriffs call it an added tool in law enforcement. In Flagler, the sheriff says that merely mentioning that a voice-stress analysis will be part of a job interview prevents some people from applying. But is that necessarily a good thing? Is a job interview — or a police interrogation — not inherently stressful?

Technology that makes or breaks the fate of suspects in police custody has unquestionable merit — so long as the merits are proven beyond reasonable doubt. Short of that, the technology is no more than a tool of deception itself.

Well said.

Department of Defense Polygraph Program Gets Makeover

Steven Aftergood reports in the Federation of American Scientists’ Secrecy News newsletter & blog that on 25 January 2007, the Department of Defense issued a new directive governing polygraph policy:

The Department of Defense has revised and supplemented its polygraph program to include non-polygraph techniques for detecting deception.

A new Pentagon directive (pdf) introduces the term “Credibility Assessment (CA),” which refers to “The multi-disciplinary field of existing as well as potential techniques and procedures to assess truthfulness that relies on physiological reactions and behavioral measures to test the agreement between an individual’s memories and statements.”

The new directive also transfers the polygraph program from the Defense Security Service to the secretive DoD Counterintelligence Field Activity (CIFA). The program will be overseen by the Under Secretary of Defense for Intelligence.

Eleven military and defense intelligence organizations listed in the directive are authorized to conduct polygraph and credibility assessment examinations.

The reliability of polygraph testing for employee screening is widely disputed on scientific grounds. But many government security officials nevertheless insist on its value and utility, and the practice persists.

See “Polygraph and Credibility Assessment Program,” Department of Defense Directive 5210.48, January 25, 2007.

Significantly, the new directive tightens control over DoD agencies’ use of any “credibility assessment” technology other than the polygraph. This seems a likely reaction to the post-9/11 debacle wherein some DoD components began using Computer Voice Stress Analysis (CVSA) to interrogate prisoners. The manufacturer of this quack device, the so-called “National Institute of Truth Verification,” has admitted in court that CVSA “is not capable of lie detection,” and the company was recently the subject of an ABC News exposé. DoD eventually put an end to its use of CVSA. The new directive ensures that henceforward, DoD agencies will use only officially approved pseudoscientific techniques for “credibility assessment” purposes.

The new directive also changes the name of the Department of Defense Polygraph Institute (DoDPI) to the “Defense Academy for Credibility Assessment” (DACA). At the time of this writing, the DoDPI website has not yet been updated to reflect this change.

ABC News Exposé of Charles Humble and CVSA on YouTube

A March 2006 ABC News Primetime story on Charles Humble, the phony Ph.D. behind the pseudoscientific Computer Voice Stress Analyzer, has been posted to YouTube:

For previous discussion of this news story, see Innocent Until Proved Guilty? (CVSA Exposé) on the message board.

Tremors of the Trade

Retired NYPD detective Warren J. Sonne writes for in an article titled, “Tremors of the Trade: Investigative Tool or Troublesome Black Magic?” Excerpt:

Over the past decade, police departments all over this country have lined up to purchase Voice Stress Analyzers. In a country that has placed restrictive rules of law on the police such as Mapp v Ohio (Fourth Amendment exclusionary rule against unreasonable searches and seizures) and Miranda v Arizona (Fifth Amendment right against self-incrimination, and Sixth Amendment right to counsel). My question is, “Are PDs playing with fire?”

I believe that the plain answer is, “yes.”

I think that the first problem comes from the “Pac-Man” generation. Give us a high-tech toy and some practice, and within an hour we will defeat those lying criminals. Once we’ve done that, we can move on to the next level. Hey, CSI solves two or three crimes during their one-hour episodes, don’t they? Well, if they can do itÂ….

There are no quick fixes to criminal investigations, nor are there magic boxes to help detectives figure out who’s lying. There are no shortcuts to competent investigations — at least, there shouldn’t be.

The second problem that I see originates in the many high profile false confession cases that have plagued our profession. The Central Park Jogger case in New York is just one such example. In April of 1989, a young woman was assaulted and raped while running inside of New York’s Central Park. Investigation identified five teenagers as suspects to this crime, and subsequent interrogations produced confessions from four of them. This led to all five being convicted for this crime. In 2002, an unrelated person confessed to this crime, and his confession was coroborrated by DNA. This person, Matias Reyes, claimed that he acted alone. Since there was no corroborating evidence against the five teenagers, other than their confessions, their convictions were overturned.

There are many other cases that have been overturned by the courts involving people, some of them mentally challenged, who have confessed to crimes that they didn’t commit. As a result, many police departments across the country have instituted polices requiring the videotaping of all confessions, with some departments recording all, or nearly all, interviews.

So, do we really need a Supreme Court decision here? Another Miranda? “You have the right to remain quiet and refuse to play the CVSA Game.”

A third issue is, does it work? What is Voice Stress Analysis (VSA), also called Computerized Voice Stress Analysis (CVSA)? The equipment now being produced is in the form of very nice looking computers that allegedly measure “micro-tremors” in the voice. The manufacturers of this equipment claim that these micro-tremors are produced when people are under stress, such as when they are lying. One obvious problem is that stress can and does occur for many other reasons in addition to lying.

Some scientific studies (not all) have verified that these micro-tremors exist and can be measured. Unfortunately, these same scientists and their studies have failed to demonstrate lie detection accuracy using CVSA at anything better than chance (flipping a coin).

I visited the web site of the leading CVSA manufacturer, The National Institute for Truth Verification, but other than their claims that CVSA works as a lie detector, I could find no scientific studies to back it up. What I did find were a half-dozen or so endorsements from police officers who were scattered around the country. Yet, not one these endorsement actually claimed that the CVSA worked as a lie detector. It was called a “valuable tool,” and most of endorsers seemed to like the training program, or the versatility of it, but not one claimed that it was an accurate lie detector.

I remember the Xerox machine being a valuable investigative tool as well. All the detective had to do was push the magic button and out came a piece of paper with the word LIE or TRUTH on it. “What’s your name?” was the first question. “Sam Jones” was the reply. Push the button and out came the answer “TRUE.” “Okay Sam, you’ve done pretty well so far. Now, did you steal your Bill Doe’s bike this morning?” asked the detective. “No,” says Sam.” Push the Xerox button, and here comes the answer, “LIE.”

Some of you may laugh, yet others will remember the utility of this less than scientific procedure. People can be awestruck by the appearance of science. Before the advent of the Xerox machine, people had to use “carbon paper” if they wanted copies. People could certainly be naïve in those days. Well, faced with the scientific appearance of CVSA, people can still be naïve in these days as well. The argument for CVSA seems to be based on the utility of the device, not the accuracy of it.

Sonne is absolutely right. But his criticisms of voice stress analysis are largely applicable not only to voice stress analysis, but to polygraphy, too.

Moscow Airport Says Passenger Screening Device Doesn’t Detect Lies: It Reads Minds

On 6 April 2006, Adrian Blomfield of the Daily Telegraph reported that Moscow’s Domodedovo Airport would soon begin requiring passengers to submit to voice-based lie detector “testing” in an attempt to identify terrorists and drug smugglers. The device, Nemesysco’s GK-1 Security Access Control System, is completely unsupported by any peer-reviewed scientific research whatsoever. Now, in an apparent attempt at public relations damage control, Domodedovo Airport’s press office has issued a press release explaining that “GK-1 is not a lie detector…. GK-1 is a system … trying to detect whether [a person] is about to commit a crime.” So it’s not a lie detector, it’s a mind reader. Got it?

Here is the full text of the press release:

MOSCOW (RNWire) – In response to some media reports, in which the GK-1 multilevel voice analysis system was erroneously referred to as a lie detector system, the press service of Domodedovo International Airport states that these systems cannot be treated as analogous. GK-1 is based on the latest computer technologies and is a further development of the traditional profiling system.

Profiling is a preliminary interviewing by trained specialists, who analyze a persons’ look, responses and behavior. This method is actively used by security services of the world largest airports and airlines in order to prevent terrorist attacks and  transportation of illegal items and substances.

GK-1 is a system automating profiling methods and increasing profiling effectiveness. This is achieved, for instance, through exclusion of the so called human factor effect, i.e. mistakes made as a result of tiredness, possible collusion, etc.

GK-1 is not a lie detector, which is focused on an important question. GK-1 is a system performing multilevel analysis of emotional aspects of a person’s voice, trying to detect whether he/she is about to commit a crime. In contrast to cases, when a crime has already been committed, the intention to commit a crime leads to a unique psychological and emotional state which can be detected by GK-1.

The checks with the use of the afore-said technology will be made selectively and under control of Domodedovo Aviation Security specialists and representative of the federal authorities. Suspicious passengers will be offered to voluntarily undergo such a test. A passenger may be denied boarding if he/she refuses to undergo a test.

GK-1 interviewing is performed in an automatic mode. Questions won’t be of personal nature and will not prejudice the person’s rights. Such interviewing is related only to prevention of terrorist attacks and transportation of illegal items and substances.

In the course of pilot operations the content and the number of questions (from three to five in accordance with the plan) may be changed in order to improve the system’s effectiveness.

The possibility and advisability of the large-scale use of GK-1 system at Domodedovo will be determined on the basis of the pilot operations’ results. The use of such a system will not cause inconveniencies to the airport visitors. On the contrary, the successful introduction of GK-1 will lead to a reduction of checks currently performed at the airport, which will improve passenger comfort.

Press Service of Domodedovo International Airport would like to draw your attention to the fact that the airport is responsible for security of all airport visitors, including passengers and people coming to meet relatives and friends at the airport. This explains the necessity to continuously improve the security system.

GK-1 multilevel voice analysis system was developed by NEMESYSCO (Israel). The company’s representative in Russia is Urbis Management Systems Software. The testing which has been carried out by Domodedovo Aviation Security, Areopag-Center, and Drug Smuggling Prevention Department of Domodedovo Customs Authorities at Domodedovo Airport for several months, confirmed that the system can be effectively used in the security zones to prevent terrorist attacks, illegal traffic of arms, ammunition, explosives, poisonous substances and drugs.

For discussion of Domodeovo Airport’s regrettable decision to field this Emperor’s New Clothes pseudo-technology, see Lie the Friendly Skies on the message board.

Lie detectors raise doubts

Northwest Indiana Times reporter Bob Kasada reports on Computer Voice Stress Analysis.

This story ran on on Saturday, April 15, 2006 12:42 AM CDT

Lie detectors raise doubts


Portage police Detective Capt. Terry Swickard has a lot of confidence in the department’s computer voice stress analyzer.

The results of the high-tech lie detector test are not admissible in court, he said, but they have helped guide investigations toward convictions and the release of innocent suspects.

As a result, Swickard was not at all swayed by news of a Pentagon-sponsored study that concluded the technology performs at chance level and is ineffective in detecting the presence of deception or stress.

The study verified earlier findings, which had led the U.S. Department of Defense to no longer use voice stress analyzers, according to a spokesperson, who said it is policy not to be identified by name.

The computer voice stress analyzer, which is among two types of units evaluated in the University of Florida study, is used by police departments in Porter and Lake counties.

The maker of the units, National Institute for Truth Verification of West Palm Beach, Fla., said in a prepared statement that this and earlier studies are flawed because they are unable to produce the same type of jeopardy experienced by those undergoing the voice test in real-life situations.

The authors of the study said they prepared for this concern by factoring in the lack of “real-world” stress as part of their evaluation.

The National Institute for Truth Verification said 1,500 law enforcement agencies have relied on the units over the past 18 years and that an independent survey of agencies reported an accuracy rate of over 91 percent.

Porter Police Chief James Spanier, who was undergoing training with the computer voice stress analyzer at the time of the Sept. 11 terrorist attacks, said there must be some consequence for lying in order for the tests to be effective.

He also said the results are only as good as the questions asked.

Defense attorney Larry Rogers said he has been aware of the criticism of the tests for several years and is not surprised by the defensive responses from police.

If police were to admit the faults with the technology, they would be opening themselves up to liability and objections over the waste of tax dollars, he said.

Each computer voice stress analyzer unit costs $10,000 and six-days of training runs $1,440 per person, said Bill Endler, director of NITV. Endler is a retired police chief from Syracuse, Ind., and has worked at other departments around the state.

Rogers said he advises all his clients to exercise their right to refuse the voice stress test.

“You might as well use a Ouija board,” he said.

Michael Higgins, spokesman for the Lake County Sheriff’s Department, said the Pentagon study will have no impact on the department, which has five certified voice stress operators.