Transport safety depends on human state.
EchoDepth makes that state visible.
Operator fatigue, passenger distress and safety communication effectiveness — the three variables that most transport operators track retrospectively, if at all. EchoDepth provides a real-time signal for all three.
What transport operators cannot currently see
Driver and operator fatigue is the leading unmeasured safety variable. You know it is a risk. You have no real-time signal for it.
Passengers in distress — through anxiety, accessibility barriers or safety incidents — cannot always self-report. The signal is in their behaviour, not their words.
Safety communication effectiveness is assumed, never measured. You issue the briefing. Nobody knows if it was understood or believed.
Neurodivergent passengers respond differently to environments and communications. Standard survey data cannot capture what facial and behavioural signal can.
Transport safety incidents attributed to operator fatigue cost the industry billions annually in liability, remediation and reputation. Passenger dissatisfaction goes undetected until it appears in surveys — retrospectively, after the damage is done.
How EchoDepth is used in transport
Operator readiness monitoring
Video (cab-facing camera)
Real-time fatigue and cognitive load score. Alert when readiness drops below threshold.
EchoDepth monitors 44 facial Action Units in train drivers, bus operators and control room staff — detecting fatigue onset before it becomes an incident factor. Integrates with existing cab monitoring infrastructure.
Passenger experience measurement
Video (station/carriage — anonymised and aggregated), feedback forms, audio
Emotional state distribution across journey stages. Distress flags for intervention.
Understand how passengers emotionally respond to delays, crowding, environmental factors and service disruptions in real time — not through retrospective surveys. Particularly valuable for accessibility and neurodiversity research.
Safety communication testing
Video of safety briefings, audio announcements, text communications
Audience engagement score and comprehension signal for safety-critical communications.
Before deploying a new safety communication — briefing video, emergency procedure, announcement — EchoDepth analyses audience emotional engagement and comprehension signals. Know whether the message landed before lives depend on it.
Accessibility and neurodiversity research
Video (with consent), journey tracking, environmental data
Emotional response mapping for neurodivergent passengers across journey stages.
Transport for Wales R&D programme explores how neurodivergent passengers emotionally respond to environments, information systems and service interactions — providing the evidence base for accessibility improvements that survey data alone cannot produce.
GDPR and public environments
Deploying emotional AI in public transport environments requires careful governance. EchoDepth operates under a consent-first architecture in all deployments.
FAQ
Yes. Audio analysis of driver communications, text analysis of passenger feedback, and image analysis of station photography all work independently of live camera feeds.
EchoDepth provides a REST API and WebSocket stream. Integration with existing operator management platforms, incident reporting systems and control room displays is supported. Contact us to discuss your specific infrastructure.
Cavefish has an active R&D partnership with Transport for Wales and Media Cymru, funded through the Welsh Government innovation programme. This is a research and development engagement exploring applications of emotional AI in transport contexts.
Working with transport operators
If you are a transport operator, infrastructure provider or safety regulator exploring emotional AI applications, get in touch.