The Platform

EchoDepth — Multimodal Emotional Decision Intelligence

44 facial Action Units. FACS standard. 14 cultural cohorts. VAD model. Real-time and video analysis. Built for enterprise deployment.

EchoDepthis Cavefish's Emotional Decision Intelligence platform. It measures and predicts emotional and cognitive responses to decisions, communications and human interactions — before those decisions are executed, before they fail, before the cost lands.

Book a Demo →How It Works

Real-Time Emotional Signal Detection

EchoDepth detects and classifies 44 facial Action Units in real time. Emotion scores update per frame — giving operators, analysts and coaches a live signal they can act on.

  • Per-frame AU detection
  • Live emotion scoring with labels
  • Attention and engagement tracking
  • Runs in-browser — no data leaves the session
EchoDepth live emotional analysis — facial tracking with real-time emotion scores
EchoDepth VAD model — Valence Arousal Dominance plotted in 3D and over time

Three-Dimensional Emotional State Mapping

The Valence-Arousal-Dominance model provides significantly more granular emotional state mapping than binary positive/negative sentiment. EchoDepth plots emotional state in 3D space — and tracks how it moves over time.

Valence

Positive vs negative emotional state

Arousal

High vs low emotional activation

Dominance

In control vs submissive state

Built on the FACS Standard. Calibrated for Enterprise.

44 Facial Action Units

Full FACS standard implementation — the most granular validated framework for emotional state measurement.

14 Cultural Cohorts

Calibrated across 14 cohorts in 6 countries. Emotion is cultural. EchoDepth is calibrated accordingly.

VAD Model

Valence-Arousal-Dominance provides three-dimensional emotional state mapping beyond simple sentiment.

Real-Time Analysis

Live facial AU detection with per-frame emotion scoring during live sessions.

Video Analysis

Frame-by-frame emotion timeline on any recorded content — presentations, interviews, calls.

Emotion Transcript

Emotional state mapped directly to transcript — see which words triggered which signals.

Enterprise API

Structured output via API for CRM, HR, governance and reporting integration.

Governance Layer

UK GDPR, ICO and sector-specific compliance documentation for every deployment.

What EchoDepth Delivers

Trust Score

Quantified trust signal in communications and interactions

Credibility Rating

Objective credibility signal independent of content quality

Resistance Indicator

Emotional resistance signal before it becomes visible behaviour

Engagement Depth

Real engagement vs surface compliance — the crucial distinction

Confidence Score

Conviction signal in decision contexts

Risk Flag

Elevated emotional risk indicator requiring intervention

Frequently Asked Questions

What is EchoDepth?

EchoDepth is Cavefish's multimodal Emotional Decision Intelligence platform. It analyses 44 facial Action Units under the FACS standard, calibrated across 14 cultural cohorts in 6 countries, using the VAD (Valence-Arousal-Dominance) model.

How many facial Action Units does EchoDepth analyse?

EchoDepth analyses 44 facial Action Units under the FACS standard — the most granular publicly validated framework for emotional state measurement.

Is EchoDepth compliant with UK GDPR?

Yes. Cavefish operates under UK GDPR and ICO frameworks (ICO registration ZB915623). All EchoDepth deployments include a governance and compliance layer.

What is the VAD model in EchoDepth?

VAD stands for Valence-Arousal-Dominance — a three-dimensional model for emotional state representation. EchoDepth uses VAD to provide significantly more granular mapping than binary positive/negative sentiment.

Does EchoDepth operate in real time?

Yes. EchoDepth supports both real-time analysis (live sessions) and retrospective video analysis, depending on deployment context.

How It WorksOutputsSecurityGovernance

Ready to See EchoDepth in Action?

Book a Demo →