Improvement of interactive EC fitting based on substitute evaluation using sound volume preference

Interactive EC Fitting is an effective fitting method to set the signal processing parameters of a hearing aid reflecting the preference of each user. However, it sometimes generates uncomfortable sounds and takes long fitting time. This paper proposes an improvement of Interactive EC Fitting by embedding a function that screens out a subset of candidate settings which are predicted to cause uncomfortable sounds by using sound volume preference. We evaluated the improved method through a simulation-based experiment. The results statistically showed that the improved method reduces the variance of evaluation values of candidate settings and accelerates the EC convergence, compared to original Interactive EC Fitting. It was therefore indicated that the improved method is effective in screening uncomfortable sounds and shortening fitting time. We will confirm the actual effectiveness of the improved method by conducting psychological experiments in our future work.