The evolution of Voice Stress Analysis — detect deception in real-time using live or recorded speech, anywhere in the world.
Detect deception using live or recorded human speech. Only a digital recorder is needed to submit data for analysis anywhere in the world, with results within seconds.
Process and display voice patterns as audio is input. Each chart is numbered and saved for easy access and review.
Works with live microphone input, digital recorders, telephone calls, and broadcast audio from television and radio.
Real-time display of both processed and unprocessed waveforms with individual word and sound isolation.
Generate hardcopy printouts on any compatible printer. Digital storage for easy case documentation and sharing.
Install on any Windows desktop or laptop. Use a hardware key to activate on multiple machines within your organization.
Analyze unstructured, free-flowing conversation — not limited to strict "Yes" or "No" exam formats like traditional polygraph.
The latest DecepTech releases introduce AI-assisted annotation, automatic speaker identification, and multi-language support — reducing examiner workload while keeping every recording fully offline and under your control.
Offline speech recognition powered by the Vosk engine automatically places word-level annotations centered over each spoken word. Supports any sample rate, stereo or mono, with no internet connection required.
Each word annotation is automatically labeled [E] for Examiner and [S] for Subject, eliminating manual speaker tagging in two-party interviews.
Dedicated Auto Annotate editions for English (GigaSpeech model) and Spanish, with automatic language-aware model selection based on application settings.
A dedicated mode for long narrative recordings disables speaker identification when examiner/subject labeling is not required — ideal for monologue interviews and statement analysis.
DecepTech checks for software updates silently at startup and displays a dismissible notification banner when a newer version is available, with a download link matched to your installed edition.
Single installer deploys the correct architecture automatically — 64-bit on 64-bit Windows, 32-bit on 32-bit Windows — for maximum performance on every platform.
DecepTech eliminates the limitations of traditional polygraph through involuntary physiological micro-tremor analysis.
| Capability | DecepTech™ VSA | Polygraph |
|---|---|---|
| Physical attachments required | None — voice only | Belts, wires, sensors |
| Exam duration | Minutes | Hours |
| Conclusive results | 99% of the time | ~66% yield conclusive |
| Inconclusive results | Never issued | Frequent |
| Can be defeated / avoided | No — involuntary response | Yes — measurable body functions |
| Drug interference | No effect | Can affect results |
| Conversation format | Free-flowing speech | Strict Yes/No only |
| Training required | 2 weeks (80 hours) | Months |
While polygraph monitors physiological markers like heart rate and blood pressure — which can be consciously avoided — DecepTech analyzes the Lippold Microtremor. This is an automatic, involuntary physiological response in the 8–14 Hz range, directly related to stress from which deception can be determined. The micro-tremor cannot be avoided or defeated.
DecepTech is effective across all interview situations including criminal investigations, pre-employment screening, insurance fraud, and anti-terrorism operations.
Order the DecepTech system and set up a training schedule for your organization.