Noise dependent Super Gaussian-Coherence based dual microphone Speech Enhancement for hearing aid application using smartphone (2001.09571v1)

Abstract: In this paper, the coherence between speech and noise signals is used to obtain a Speech Enhancement (SE) gain function, in combination with a Super Gaussian Joint Maximum a Posteriori (SGJMAP) single microphone SE gain function. The proposed SE method can be implemented on a smartphone that works as an assistive device to hearing aids. Although coherence SE gain function suppresses the background noise well, it distorts the speech. In contrary, SE using SGJMAP improves speech quality with additional musical noise, which we contain by using a post filter. The weighted union of these two gain functions strikes a balance between noise suppression and speech distortion. A 'weighting' parameter is introduced in the derived gain function to allow the smartphone user to control the weighting factor based on different background noise and their comfort level of hearing. Objective and subjective measures of the proposed method show effective improvement in comparison to standard techniques considered in this paper for several noisy conditions at signal to noise ratio levels of -5 dB, 0 dB and 5 dB.