Prediction of binaural speech intelligibility when using non-linear hearing aids.

Abstract

A new objective measurement system is proposed to predict speech intelligibility in binaural listening conditions for use with hearing aids. Digital processing inside a hearing aid often involves non-linear operations such as clipping, compression, and noise reduction algorithms. Standard objective measures such as the articulation index, the speech intelligibility index (SII), and the speech transmission index have been developed for monaural listening. Binaural extensions of these measures have been proposed in the literature, essentially consisting of a binaural pre-processing stage followed by monaural intelligibility prediction using the better ear or the binaurally enhanced signal. In this work, a three-stage non-linear extension of the binaural SII approach is introduced consisting of (1) a stage to deal with non-linear processing based on a simple signal separation scheme to recover estimates of speech and noise signals at the output of hearing aids [Hagerman and Olofsson, Acust. Acta Acust. 90, 356 (2004)], (2) a binaural processing stage using the equalization-cancellation model [Beutelmann and Brand, J. Acoust. Soc. Am. 120, 331 (2006)], and (3) a stage for intelligibility prediction using the monaural SII [ANSI-S3.5, 1997 (R2007)]. Details of the new procedure will be discussed. [Research supported by NSERC (Canada).]