Prosodic feature recognition for threat detection

Main investigator:
Prof. Dr. Stefan Wermter
Universität Hamburg

Collaboration partners:

  • Fondazione Istituto Italiano Di Tecnologia
  • Honda Research Institute Europe GmbH

Competence Area: Situation


In an interaction, cues for detecting threatening situations could be changes in the intonation or loudness indicating dangerous situations. Robots should be able to perceive and react to such signals and change their behaviour to avoid the threatening situation or, in case the threat cannot yet be identified, be alert and act more carefully. In order to achieve this goal, this project aims at identifying prosodic features using robust neural sound localisation in order to change the robots behaviour or trigger the inclusion of other sensory pathways.

Expected Results

The outcome of this project will be a new learning approach integrating perception of prosodic features to identify and avoid possibly threatening situations by subsequently changing the robot’s behaviour or trigger the inclusion of other sensory pathways to identify the threat.