ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Binary Mask for Speech Enhancement

Journal: INTERNATIONAL JOURNAL OF ELECTRONICS & DATA COMMUNICATION (Vol.2, No. 1)

Publication Date:

Authors : ; ;

Page : 1-3

Keywords : Ideal binary masking; Signal to noise ratio; time-frequency; Automatic speech recognition; voice activity detector; Minimum mean square error;

Source : Download Find it from : Google Scholarexternal

Abstract

This paper is based on general principle of ideal binary masking. To Improve Intelligibility of the speech Ideal binary masking is used based on noise distortion constraints. A binary mask is designed to retain noise overestimated T-F units while discarding noise underestimated T-F units. Listening tests were conducted to evaluate the new binary mask in terms of intelligibility. An ideal binary mask is a priori defined as a binary matrix where 1 indicates that the target is stronger than the interference within the corresponding time- frequency unit and 0 indicates otherwise. Missing feature methods have been shown to be very successful at compensating for the effects of stationary and non-stationary noise when this mask is computed from a priori knowledge of the SNR of all spectrographic components. Our study, thus, demonstrates that the use of binary masking represents a promising direction for speech enhancement.

Last modified: 2016-07-04 17:33:22