【摘要】 现有的语音盲分离算法都是针对说话人个数已知且在分离过程中恒定的情况,现实中说话人个数是时变的.提出仅用2个麦克风对说话人个数时变的语音信号进行盲分离的方法,不仅能反映说话人个数时变的过程,而且解决了麦克风与说话人之间个数差异引起的超定、欠定问题.更多还原

【Abstract】 Most algorithms available of blind source separation in speech are limited to the situation that the number of speakers is known and constant during the process of separation.But in reality the number of speakers is time-varying.This paper uses only two microphones to separate speech signals of dynamic speakers.The method can not only reflect the change of number of speakers but also solve the overdetermination and underdetermination problems caused by the variation of number of speakers and microphones.更多还原