New idea of hearing aid algorithm to enhance speech discrimination in a noisy environment and its experimental results

Abstract

A hearing aid can help the hearing impaired to listen to speech by amplifying its sound. However, amplification cannot always improve speech perception. To improve speech perception in a noisy environment, we proposed a new hearing aid algorithm that combines independent component analysis (ICA) with multiband loudness compensation. The proposed method was applied to a behind-the-ear type (BTE) dual microphone hearing aid. The proposed method successfully separated a target speech signal from background noise such as babble, car noise, factory noise. This method may help hearing aid users to hear the speech that he or she wishes to hear clearly in noisy conditions.