Neumorphic acoustic sensor systems for next-generation hearing aids
From March 24 to 26, 2025, the NeuroSensEar consortium met for its spring meeting at the University of Ulm to discuss progress in the development of neumorphic acoustic sensor systems. The aim of the project is to create innovative technologies for energy-efficient, context-aware hearing aids of the future - inspired by biological models and supported by novel hardware solutions.
The participation of three members of the Advisory Board was particularly pleasing: Prof. Christian Dobel (ENT Clinic Jena), Prof. Bastian Epp (Technical University Denmark, expert in coupled oscillators in hearing and psychoacoustics) and Prof. Eckehard Schöll (TU Berlin & Potsdam Institute for Climate Impact Research).
Opening and scientific impulses
The project kicked off on Monday afternoon with a presentation of the overall project by project coordinator Prof. Claudia Lenk (Institute for Functional Nanosystems, AG Biomedical Sensors and Microsystems, University of Ulm), followed by an inspiring keynote speech by Prof. Dr. Heiko Neumann (Institute for Neuronal Information Processing, University of Ulm). In his lecture "Contextual control for binding in vision and audition - from neuroscience to neuromorphic computing", he traced the arc from biological perception to technical implementation in neuromorphic systems.
After a joint poster session and the subsequent exchange over dinner, the full scientific program continued on Tuesday.
Specialist sessions and interdisciplinary exchange
Maurits Ortmanns (Institute of Microelectronics, University of Ulm) opened Tuesday with an overview of "Microelectronics @ UUlm - Implantable Electronics for Neural Recording, Stimulation and Modulation". Current research topics from the project were then presented in four exciting sessions:
- MEMS cochlear hardwareProgress in microelectromechanical cochlear components and their integration into electronic systems.
- Insect hearing as a modelApproaches to replicate biological localization mechanisms.
- Speech intelligibility and sound quality: Models and tests for acoustic perception.
- Acoustic scene analysis: Use of deep neural networks and reservoir computing for pattern recognition in complex sound environments.
Presentations for the consortium were given by
- Prof. Manuela Nowotny (FSU Jena)
- Fan Zhang (TU Ilmenau)
- Dr. Ilko Pieper (TU Ilmenau)
- Dr. Jakob Abeßer (Fraunhofer IDMT)
- Alireza Molla Ali Hosseini (TU Ilmenau)
The medical context was not neglected either: in her keynote speech "More Than Hearing: The Hidden Complexity of Auditory Perception", Prof. Dr. med. Anna-Katharina Rohlfs (Department of Phoniatrics & Pedaudiology, Ulm University Hospital) impressively demonstrated how multi-layered human auditory perception is. The scientific conclusion of the day was the lecture by Prof. Dr. Marc Ernst (Institute of Applied Cognitive Psychology), who dealt with the integration of sensory impressions for robust spatial hearing.
The intensive day ended with internal project discussions.
Conclusion with a focus on AI
On Wednesday, the workshop moved to the Launchpad at Ulm University - an innovative seminar room specially designed for creative, interactive teaching formats.
In this environment Prof. Dr. Daniel Alexander Braun opened the day with his keynote "Continual Reinforcement Learning with Information Constraints" - an exciting outlook on AI systems that can learn continuously and flexibly. This was followed by internal discussions and a poster session before the workshop ended with a joint lunch.
