Is hearing-aid signal processing ready for machine learning?


  • Bert de Vries GN ReSound, Eindhoven, Netherlands Eindhoven University of Technology, Eindhoven, Netherlands
  • Andrew Dittberner GN ReSound, Glenview, IL, USA


In the hearing-aids community, machine-learning technology enjoys a reputation as a potential performance booster for signal-processing issues such as environmental steering, personalization, algorithm optimization, and speech detection. In particular in the area of in situ hearing aid personalization, the promise is steep but clear success stories are still hard to come by. In this contribution, we analyze the ‘personalizability’ of typical hearing-aid signal-processing circuits. We discuss a few salient properties of a very successful adaptable and personalized signal-processing system, namely the brain, and we discover that among some other issues, the lack of a probabilistic framework for hearing-aid algorithms hinders interaction with machine-learning techniques. Finally, the discussion leads to a set of challenges for the hearing-aid research community in the quest towards in situ personalizable hearing aids.


Friston, K. (2009). “The free-energy principle: a rough guide to the brain?” Trends Cogn. Sci., 13, 293-301.

Jaynes, E.T. (2003). Probability Theory: The Logic of Science. Cambridge University Press.

Kochkin, S. (2010). “MarkeTrak VIII: Consumer satisfaction with hearing aids is slowly increasing,” Hearing Journal, 63, 19-20,22,24,26,28,30-32.

McCandles, D. (2010). “The beauty of data visualization”, Talk at TED Global Oxford ( Based on work by Tor Nørretranders.

ON Semiconductor (2013). “Datasheet for AYRE SA3291 Preconfigured DSP System for Hearing Aids,”

Raskin, A. (2011). Wanna Solve Impossible Problems? Find Ways to Fail Quicker.




How to Cite

de Vries, B., & Dittberner, A. (2013). Is hearing-aid signal processing ready for machine learning?. Proceedings of the International Symposium on Auditory and Audiological Research, 4, 121–132. Retrieved from



2013/2. Physiological correlates and modeling of auditory plasticity