Assessing Deception by Voice Analysis: Part II: The LVA
This overview is the second of a two-part series designed to brief law enforcement personnel, attorneys and members of the intelligence services about the ability of voice analyzers to detect deception and stress from speech. Such analyses are important as devices of this type are proliferating and their use is creating problems. Thus, the question was asked: Can these devices actually provide valid information about deceptive behaviors? The effectiveness of NITV’s Computer Voice Stress Analyzer (CVSA) was reviewed in the Part I article (Hollien and Harnsberger, 2013) where it was reported that the system was found to be ineffective. This second of the two reviews will focus on a somewhat different device – Nemesysco’s Layered Voice Analyzer (LVA). Here the summary will provide information about 1) its background, 2) relevant research, 3) a large laboratory experiment and 4) an extensive field study. The highly controlled laboratory experiment, assessing LVA, employed speech samples of individuals who systemically varied their utterances from normal to those which were intensely deceptive. To create the latter, subjects had to hold very strong views about some issue and were required to make sharply derogatory statements about them while believing that they would be observed by colleagues and friends. A double-blind evaluation of these utterances was carried out by two teams of qualified LVA operators. The field experiment (Horvath et al 2013) focused on a group of suspects split in two groups -- a group that had produced truthful utterances and another whose members produced deceptive speech. The veracity of these utterances was validated by several procedures based primarily on polygraph evaluations. The detection rates provided by LVA operators were both assessed directly and then contrasted with the accuracy of experienced auditors. The results obtained by both studies demonstrated that the LVA system operates at only about chance levels.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).