Professionals in the field of truth verification know there is no such thing as a real “Lie Detector”, as lying is not associated with a single specific mental state, and human reactions are highly dependent on the test context. InTone.ai Risk is an emotion detector, identifying reactions in the tested party’s voice in response to pre-scripted carefully designed sets of questionnaires. Its output, the “Risk Level” detected per topic, is calculated based on plurality of indications surrounding the specific topic, which makes it stand out from the rest of the test. Such unique and intensive reactions typically indicate a problem that should be further explored by a professional investigator.

InTone.ai Risk analysis technology assumes that often in this use some emotional states are not indicative of “risk” and are indeed a result of the true unpleasant situation – therefore, InTone.ai Risk uses a combination of techniques designed to “clean” the analysis and normalize it, so only the truly problematic items that stand out will be identified and reported to the operators.

Absolutely. InTone.ai Risk has multiple user interfaces designed to provide each user with the information he needs to complete his task. Using the scripted protocol system InTone.ai Risk exposes through its call center level interfaces, it knows how to associate its detection to specific topics and questions and to automatically generate a report your investigation team (SIU) can use in InTone.ai Risk through their professional work to reach a final decision about the case.

Yes, InTone.ai Risk systems can be configured to support almost any organization structure you may have. In case of file transition, all relevant files (audio and results) are zipped, encrypted and sent via your own FTP account to their destination.

Having the basis for a proper fraud prevention system in place is a great entry point for RA7, and InTone.ai Risk can be monetized differently in different organizations. InTone.ai Risk can reduce the number of cases sent for outsourced investigations by identifying Low Risk cases (saving time and hassle for your honest client!), it can shorten the time and minimize the resources put in each investigation, leading them directly to the “soft points’; in the story, and of course it can help identify and clear many more soft-fraud cases, even for low value claims, during the first phone call.

InTone.ai Risk accuracy is dependent upon several factors:

1. First and foremost, the emotional involvement and jeopardy of the tested party with regard to what is being analyzed. Jokes and lies of minimal involvement will produce very limited emotional reactions, and the calculation of risk will rarely show indications on such jokes.
2. The quality of the input and signal to noise ratio. High noise levels, bad phone connection, too low or too high input volumes will produce erroneous results.
3. The test settings and possible uncontrolled interferences or environmental distractions.
4. “Borderline” personality issues or clinical conditions of the tested party may cause strange or inaccurate results.

Essentially, if the quality of the voice is reasonably good and the operation and preparation is proper, the emotional analysis component will be extremely accurate. If the intention to deceive or conceal the truth is genuine and poses jeopardy on the tested subject, then the Risk Calculations will also be very accurate.