Medical Technology Systems - Project "AudEeKa"


Auditory emotion recognition for empathic AI assistants (AudEeKa)

The AudEeKA project was launched with the aim of developing an empathic AI assistant. With the help of such assistants, the consideration of so-called "soft factors" in medicine is to be strengthened, because it is precisely the emotional level that represents an important component in the relationship between patient and doctor. Since this is becoming less of a focus with the increasing use of telemedical applications, but also with the ongoing tense situation in health care, in future AI-based assistance systems should not only support diagnoses and therapy options, but also increasingly emphasise the emotional situation.

Methods for emotion recognition based on audio data are initially being developed for the assistant, but it should not only be able to recognise emotions via speech, but ideally also use a wide variety of biosignals and stress markers to classify the emotional state. In order to make the assistant as mobile as possible, one of the main focuses of development is usability in low-resource environments.

Project data

Duration: 01.11.2022 – 31.10.2025
Donee: University Duisburg-Essen/ Medical Technoloy Systems
Sonsor: Bundesministerium für Wirtschaft und Klimaschutz
Deutsches Zentrum für Luft- und Raumfahrt e.V.
Gran number: Founded by the Federal Ministry for Economic Affairs and Climate Action, German Aerospace Center e.V. FKZ: 50RP2260A (UDE) und 50RP2172B (LMU)
Partner: Ludwig-Maximilians-Universität (LMU) München, LMU Klinikum, Klinik für Anästhesiologie


Prof. Dr. rer. nat. Elsa Kirchner

Contact person:
Roswitha Duwenbeck, M.Sc.