On Wednesday, 9 November 2011, CBS Atlanta aired an interview with private investigator T.J. Ward, who used a computer program to analyze recorded statements by Republican presidential candidate Herman Cain and Sharon Bialek, who alleges that Cain made a sexual advance on her when she sought his help getting a job. According to Ward’s software, Cain “is being truthful, totally truthful” in denying Bialek’s claim, while Bialek “is fabricating what transpired.”
Ward claimed his software cost $15,000 and has an accuracy rate of 95%. While CBS Atlanta did not disclose the name of this software, a link on Ward’s website indicates that it is Layered Voice Analysis (LVA), a program developed by Amir Liberman’s Nemesysco, Ltd., an Israeli company. This software was completely discredited by Swedish linguists Anders Eriksson and Francisco Lacerda in a 2007 article titled “Charlatanry in Speech Science: A Problem to Be Taken Seriously” (International Journal of Speech, Language and the Law, Vol. 14, No. 2). For more on LVA, see “Nemesysco Founder Amir Liberman Is a Charlatan.”
Not that it matters (LVA being pseudoscientific flapdoodle), but it is perhaps worth observing that private investigator T.J. Ward had chosen the setting for analyzing a male voice when conducting his analysis of Sharon Bialek’s remarks:
CBS Atlanta should have done its homework before running with a “news” story based on emperor’s-new-clothes technology.
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and Eriksson and published in the International Journal of Speech, Language and the Law in December 2007, but later withdrawn.
In fact, the article has not been withdrawn, rescinded, or otherwise disavowed by the journal that published it.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.
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.
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.