The purpose of this study was to look into the statistical aspects of the formant-modulation approach for determining speech intelligibility. The proposed formant-modulation approach is described in great detail. The expectation and variance of estimators of a modulation coefficient and effective signal-to-noise ratio are calculated using analytical equations. A major noise disturbance and a predominant reverberant disturbance are two circumstances that are particularly intriguing in engineering applications. It has been discovered that engineering applications equations can be used to find efficient test signal duration for needed measurement exactness. The formant-modulation method’s efficacy has been proven by computer modelling and experimental testing. The results of a comparison of the formant and formant-modulation methods of assessing speech intelligibility led to the conclusion that the formant method is preferred in cases where noise is present. Formant method estimate time is around 1/14 of formant-modulation method estimation time, and relative error is roughly 0.15-0.20 of formant-modulation method error. However, with a large effect of reverberation on speech intelligibility, the formant approach cannot compete with the formant-modulation method.
Department of Acoustic and Multimedia Electronic Systems, National Technical University of Ukraine ‘Igor Sikorsky Kyiv Polytechnic Institute’, 37, Prosp. Peremohy, Kyiv, Ukraine.
View Book:- https://stm.bookpi.org/IST-V1/article/view/5083