Multimodal Deception
Detection

Real-time analysis of speech, voice, micro-expressions and behavioural patterns during form filling – with zero personal data stored.

Capture

Analyze thousands of behavioral signals

Detect inconsistencies

Support decision

Document & Audit

How Veriteus Works

Form Module

Behavioral interface analysis

Detects hesitation, text edits, typing speed, answer order and cognitive load patterns during digital form completion.

Audio Module

Real-time voice analysis

Linguistic patterns, voice stress markers and prosodic features – developed with Prof. Julia Hirschberg (Columbia University).

Video Module

Micro-expression & body signal analysis

Facial cues, gaze patterns, gesture inconsistencies and physiological indicators captured via standard webcam.

Modules can be used independently or combined for maximum accuracy and flexibility.

Up to 40 %

reduction in fraud losses

based on internal benchmarks and real-world insurance use cases

60–80 %

faster investigation times

eliminating manual review of non-suspicious cases

Zero PII

captured or stored

fully compliant with GDPR & EU AI Act high-risk classification

Backed by world-leading experts and institutions

Insurance Claims

Detect fraudulent claims before payout

Corporate Compliance

Internal investigations & ethics hotlines

HR & Recruitment

Background verification & screening

Legal & Forensics

Evidence gathering & witness analysis

Healthcare

Patient history & insurance fraud prevention

Public Sector

Benefits fraud & regulatory compliance

The Science Behind Veriteus

Not intuition. Not polygraphs. Science-based behavioral AI.

Our algorithms analyze micro-patterns in language, voice, and behavior that humans can’t reliably detect. Every signal is cross-referenced, every result is traceable, and every decision is yours.

Behavioral AI

Not guesswork. Machine learning trained on thousands of verified interactions

Peer-Reviewed Methods

Based on decades of research in linguistics, psychology, and data science

Transparent Metrics

Clear confidence scores, explainable results, audit-ready reports

Why Veriteus

Traditional Voice Stress Analysis systems

Veriteus (multimodal behavioral analysis)

Behavioral interface analysis

38–53%*

84 %

Active false positive prevention

No

Yes

What the method specifically detects

Primarily voice micro-tremor (stress), does not distinguish natural stress/anxiety from deception

Linguistic patterns, acoustic markers, micro-expressions, typing behavior – detects manipulation & inconsistency

Real-time analysis

No

Full real-time across all modules

Highlighting suspicious phrases

Only overall “stress score” or binary truth/lie output, no detail

Fully explainable – highlights specific phrases, words, micro-expressions with confidence scores

Data processing

Usually requires storing voice recordings (PII risk)

Zero PII – nothing stored, only temporary processing

Modalities

Audio only (voice)

3 independent modules: Form + Audio + Video (standalone or combined)

* Reid. (2002). The Computerized Voice Stress Analyzer (CVSA). https://reid.com/resources/investigator-tips/the-computerized-voice-stress-analyzer-cvsa; Elkins, A. C. (2010). Evaluating the credibility assessment capability of vocal analysis software. Proceedings of the 43rd Hawaii International Conference on System Sciences (HICSS), 1–9. https://ministryoftruth.me.uk/wp-content/uploads/2012/02/elkins2010.pdf; Horvath, F., McCloughan, J., Weatherman, D., & Slowik, S. (2013). The accuracy of auditors’ and layered voice analysis (LVA) operators’ judgments of truth and deception during police questioning. Journal of Forensic Sciences, 58(2), 385–392. https://doi.org/10.1111/1556-4029.12066

Ready to see Veriteus in action?

Typical demo lasts 20 minutes and is tailored to your exact use case (insurance claims, KYC, HR screening, investigations).