A2I Augmented Auditive Intelligence


The dynamics of complex diseases and the effectiveness of interventions can often only be followed through continuous monitoring of multiple vital parameters. The merging of sensory vital data offers great potential for the use of artificial intelligence to improve medical decision support and to determine early warning signals that otherwise often remain hidden. At the same time, the recorded vital data enable a reduction in listening fatigue in difficult listening situations with several simultaneously active speakers by amplifying the speech signal of the speaker on whom the user is concentrating.


The aim of the project is the user centered development of a wearable multi-sensor system based on a networked hearing aid that fuses additional sensors for measuring cardiovascular data, EEG sensors for measuring brain activity, temperature and movement sensors and enables complex medical diagnoses and decisions based on artificial intelligence processes (AI). The ear provides a location that is suitable for precise long-term measurements, which allows a hearing system equipped with sensors to be integrated into the everyday life of the patient. Together with patients and medical professionals, concepts for using the system to improve processes in the field of audiology, occupational medicine and cardiological care are developed and tested.


The use of artificial intelligence processes in connection with the increasing miniaturization of vital data sensors opens up new possibilities for prognosis and diagnosis of health conditions, which can, however, radically change the doctor-patient relationship. The design of the cooperation between man and machine and the transparency of decision-making are of decisive importance in this context.


External Leader

Dr. rer. nat. Marco Eichelberg
Non-Intrusive Binaural Speech Intelligibility Prediction From Discrete Latent Representations

A. F. McKinney and B. Cauchi; IEEE Signal Processing Letters; 2022

Non-intrusive prediction of speech intelligibility for the first Clarity Prediction Challenge (CPC1)

A. F. McKinney and B. Cauchi; Clarity Challenge; 0December / 2022

Predicting Recovery from Coma After Cardiac Arrest Using Low-level Features from EEG Recordings and a Small-sized LSTM Network

Benjamin Cauchi, Marco Eichelberg and Andreas Hein; Computing in Cardiology; 0October / 2023

Kompetenzzentrum HörTech gGmbH
Advanced Bionics GmbH
Carl von-Ossietzky-Universität Oldenburg, Fakultät für Medizin und Gesundheitswissenschaften, Abteilung Neuropsychologie
Iconstorm Next GmbH & Co KG
Herz- und Diabeteszentrum NRW
Christian-Albrechts-Universität zu Kiel, Institut für Innovationsforschung, Lehrstuhl für Technologiemanagement


Start: 01.03.2021
End: 29.02.2024

Source of funding

Related projects


Development of Product Service Systems in Tele Audiology