Present speech separation methods are mostly built on the speech mixing in the noise-free situation, and are primarily classified in two large categories: Blind Source Separation (BSS) and Computational Auditory Scene Analysis (CASA).

• 目前的混叠语音分离大多是建立在无噪环境中的混叠情形下，主要以盲源分离(Blind Source Separation，BSS)和计算声场景分析(ComputationalAuditory Scene Analysis，CASA)两类方法为主，前者是根据信号的统计特性从几个观测信号中恢复出未知的独立源成分；