Nemesysco Controversy Roundup

Israeli lie detector company Nemesysco has issued a press release responding to  Professors Anders Eriksson and Francisco Lacerda’s article, “Charlatanry in Speech Science: A Problem to Be Taken Seriously,” which laid bare in devastating detail the pseudoscientific nature of Nemesysco’s lie detection “technology.” It should be noted that Nemesysco’s press release opens with a misleading characterization of Erksson & Lacerda’s article:

We wish to clarify our position with regard to the so-called ‘scientific research’ written by Professors Lacerda and Eriksson and published in the International Journal of Speech, Language and the Law in December 2007, but later withdrawn.

In fact, the article has not been withdrawn, rescinded, or otherwise disavowed by the journal that published it. Rather, the publisher (rather cravenly, in our opinion) withdrew the on-line availability of the article in response to legal threats from Nemesysco’s lawyers. For background, including part of the correspondence between Nemesysco and the publisher, see Nemesysco Founder Amir Liberman Is a Charlatan on this blog.

The “Ministry of Truth” blog, which has been following the  saga particularly as it pertains to the use of Nemesysco’s lie detection software in the United Kingdom (where it is marketed as “Voice Risk Analysis”), provides a point-by-point critique of Nemesysco’s press release.

See also Professor Lacerda’s 12 March 2009 blog entry, LVA-technology and Nemesysco’s official statement, in which he responds to an earlier released statement by Nemesysco in response to his and Professor Eriksson’s article. Continue reading Nemesysco Controversy Roundup

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

“Could Your Voice Betray You?”

Douglas Heingartner reports for the New York Times. Excerpt:

IT is a time-honored interrogation tool and a staple of film noir: the lie-detector test that can incriminate or exonerate.

But such tests need not involve strapping someone to a machine. In fact, they may not require the subject’s presence – or awareness – at all. And their use is growing far beyond criminal investigations.

Increasingly, lie-detector tests use voice stress analysis, a technology that has been around for decades but that has gained in popularity as the software at its heart continues to be refined.

“It can really be done anywhere,” said Detective Pat Kemper of the Springfield Township Police Department in Ohio, who says he has used the voice-based testing to question thousands of suspects over the last decade. “It can be done via a telephone recording. It can be done covertly. You can use it for anything.”

Indeed, beyond its applications in law enforcement, proponents of the voice-based technology see its utility in everything from telemarketing to matchmaking. In Britain, a growing number of insurance companies have been using it to screen telephone claims in hopes of rooting out fraud – a goal they say has been borne out, both in fraud detection and in deterrence. One insurer, Admiral, says 25 percent of its car-theft claims have been withdrawn since it began using the system a year ago.

But the technology’s reliability is still a matter of debate, and its migration from the interrogation room to the call center has raised concerns about potential privacy implications. Voice analysis of this nature, after all, can easily be conducted without the speaker’s knowledge. Now that it is being used in the insurance industry, for example, the concerns include how a suspect claim might affect a customer’s subsequent applications for insurance.

The makers of one system, known as the Computer Voice Stress Analyzer, try to address such worries by controlling access to the technology. “We only sell to law enforcement and the government,” said David Hughes, executive director of the National Institute for Truth Verification, the company selling the system.

The analyzer has been available since 1988, and the company says it is used by over 1,400 law enforcement agencies across the United States, as well as by other state and federal agencies including the Defense Department. “In private industry, you’re doing it for profit, so there’s always a concern about ethics violations,” Mr. Hughes said. “You can’t be cavalier.”

But that distinction does not appease everyone. “Government is not necessarily more responsible than private industry,” said Bob Barr, a former congressman from Georgia who is a privacy consultant for several organizations. “The government can put you in jail if it doesn’t like what it finds.”

Actually, the reliability of voice stress analysis is not a matter of debate amongst scientists. The claims of voice stress analysis proponents are entirely unsupported by any peer-reviewed research whatsoever.

“Lie-Detector Glasses Offer Peek at Future of Security”

R. Colin Johnson reports for the EE Times. Excerpt:

Portland, Ore. — It may not be long before you hear airport security screeners ask, “Do you plan on hijacking this plane?” A U.S. company using technology developed in Israel is pitching a lie detector small enough to fit in the eyeglasses of law enforcement officers, and its inventors say it can tell whether a passenger is a terrorist by analyzing his answer to that simple question in real-time.

The technology, developed by mathematician Amir Lieberman at Nemesysco in Zuran, Israel, for military, insurance claim and law enforcement use, is being repackaged and retargeted for personal and corporate applications by V Entertainment (New York).

“Our products were originally for law enforcement use — we get all our technology from Nemesysco — but we need more development time [for that application],” said Dave Watson, chief operating officer of parent V LLC ( “So we decided to come out sooner with consumer versions at CES.”

The company showed plain sunglasses outfitted with the technology at the 2004 International CES in Las Vegas earlier this month. The system used green, yellow and red color codes to indicate a “true,” “maybe” or “false” response. At its CES booth, V Entertainment analyzed the voices of celebrities like Michael Jackson to determine whether they were lying.

Besides lie detection, Watson said, the technology “can also measure for other emotions like anxiety, fear or even love.” Indeed V Entertainment offers Pocket PC “love detector” software that can attach to a phone line or work from recorded tapes. It’s available for download at Instead of color-coded LEDs, a bar graph on the display indicates how much the caller to whom you are speaking “loves” you. V Entertainment claims the love detector has demonstrated 96 percent accuracy. A PC version is due next month.

The heart of Nemesysco’s security-oriented technology is a signal-processing engine that is said to use more than 8,000 algorithms each time it analyzes an incoming voice waveform. In this way it detects levels of various emotional states simultaneously from the pitch and speed of the voice.

The law enforcement version achieved about 70 percent accuracy in laboratory trials, according to V Entertainment, and better than 90 percent accuracy against real criminal subjects at a beta test site at the U.S. Air Force’s Rome Laboratories.

“It is very different from the common polygraph, which measures changes in the body, such as heart rate,” said Richard Parton, V’s chief executive officer. “We work off the frequency range of voice patterns instead of changes in the body.” The company said that a state police agency in the Midwest found the lie detector 89 percent accurate, compared with 83 percent for a traditional polygraph.

See the message board discussion thread, Layered Voice Analysis (LVA) for related discussion.