King-Kopetzky syndrome: An approach for a solution

摘要

King-Kopetzky syndrome (KKS) is recognized as a clinical disorder in the hearing system of a human being. It is defined as the condition in which an individual possess a normal hearing threshold on pure tone audiometry however complains about his inability or difficulty in understanding speech in the presence of background noise. In what follows is a proposed approach for a solution for KKS consisting of three main blocks starting by separating mixed signals, then identifying the speaker of each speech signal, and finally applying a selectivity method for selecting the intended speaker. Convolutive Blind Source Separation (CoBliSS) was used for signal separation while speaker identification depended on Mel Frequency Cepstral Correlations as features and Multilayer Perceptron (MLP) as a classifier. The results show that the proposed technique can be used as a possible solution for KKS.