Rethinking Hearing Aids as Recommender Systems

The introduction of internet-connected hearing aids constitutes a paradigm shift in hearing healthcare, as the device can now potentially be complemented with smartphone apps that model the surrounding environment in order to recommend the optimal settings in a given context and situation. However, rethinking hearing aids as context-aware recommender systems poses some challenges. In this paper, we address them by gathering the preferences of seven participants in real-world listening environments. Exploring an audiological design space, the participants sequentially optimize three audiological parameters which are subsequently combined into a personalized device configuration. We blindly compare this configuration against settings personalized in a standard clinical workflow based on questions and pre-recorded sound samples, and we find that six out of seven participants prefer the device settings learned in real-world listening environments.