Objective Japanese intelligibility prediction for noisy speech signals before and after noise-reduction processing

In this paper, we investigate eight objective speech intelligibility prediction measures for noisy signals before and after noise-reduction processing in Japanese. The Japanese speech signals were first corrupted by three types of noises at two signal-to-noise ratios and processed by four classes of noise-reduction algorithms, whose intelligibility was subsequently predicted by objective measures. The subjective intelligibility ratings were obtained through a set of listening tests. Further investigations were conducted for objective measures in predicting speech intelligibility of noise-reduced signals between subjective evaluation scores and objective prediction results, and of noisy signals before and after noise-reduction processing, in terms of correlation analysis and prediction errors.