Neurofuzzy approaches for advance hearing devices

Abstract

An important issue in the use of hearing devices, is estimating the specific set of gains requirements, referred to as the "target". Conventional trial and error methods are inefficient. This paper presents a neurofuzzy approach for fitting hearing aid parameters. A neural network is designed to generate the initial target and fuzzy logic is used in fine-tuning the initial target. The introduction of the neurofuzzy approach is expected to accelerate the fitting process. It may also enable audiologists to incorporate their knowledge into the system and make it fit specific needs.