Vocal expression tracked over time. 

This enables the detection of phases of high and low emotional energy. Viewed in context, this allows particularly stressful and depressive phases of the driver to be analyzed.

The Mood Diary can be used independently by the person driving and serve as a basis for decision-making in certain situations: 

• Route planning – especially for long and stressful journeys
• Decisions about driving assistance – physical and technical
• Seeking professional help in the case of strongly negative phases

Voice analysis for anger management: 

Anger tempts people to drive too fast because this expression increases risk-taking. An angry expression leads to faster speech, more precise articulation, and fewer pauses. Resulting in quasi-audible driving behaviour. Our solution addresses this by:

• Proposing breaks
• Playing certain music selection – storage in Mood Diary to verifiably test the calming effect on the driver
• Adapting car voice to driver‘s voice (shown to reduce accidents: Nass et. al, 2005)
• Authentic feedback to sensitizes and create empathy towards other drivers (Roidl et. al, 2014)

Kids on the backseat:

When children get loud, the risk of driver distraction and cognitive overload is increased. To control this, the voice assistant:

• Offers automatic entertainment program for backseat screens
• Proposes childrens favourite music or radio play
• Engages children in games, voice chat with avatars, etc.

Significant changes in the user’s voice can be tracked through interactions with voice assistants. Voice analysis can be used for health monitoring to identify vocal biomarkers.

Possible vehicle actions:

• Suggestion to call doctor for appointment or check-up
• Emergency calls – hospital, doctor or emergency contact
• Emergency stop & warnings in autonomous driving, 

Identification of dangerous situations in the car:

• Detection of alcohol intoxication of driver or side effects of strong medication
• Automatic detection of violence and aggression by drivers and passengers- car sharing, taxis, etc.

Possible vehicle actions:

• Autonomous driving, emergency stop & warnings for other drivers
• Silent alarm to police or taxi operator in case of violent passengers

Multi-model sleepiness detection enhanced by audio:

• Non-speech sounds: yawning, snoring
• Analyzed speech from voice prompts and telephone calls

Possible action points:

• Suggestion to take a break at nearby restaurants
• Changing climate settings, playing sounds & selecting upbeat music
• Autonomous driving, emergency stop & warnings for other drivers

Enhancing driver attention control systems with audio data :

General detection of cognitive load for better stress management.

• Detects if driver is distracted by a conversation over the phone or with passengers
• Measuring stress or high cognitive load in speech (irregular speech rhythm, pauses, repetitions, hesitations, high arousal, etc.)

• Detecting sound events that might indicate or cause driver stress (children crying, objects falling, etc.)

Possible vehicle actions:

• Adaptation of the voice assistant interactions (simplifying conversation, lowering speaking rate, pausing or auto-answering phone calls)

• Autonomous driving, emergency stop & warnings