“…The objective of this study is to examine the feasibility of using a Zero frequency filter (ZFF) based system to address the streaming level. Since ZFF was developed for the analysis of speech, it has been widely employed for both epoch extraction [22]- [23] and pitch estimation [24]. Recently, Yegnanarayana and Prasanna [25] briefly proposed a discriminative technique to separate speaker excitation features in a two-speaker mixture, then use it to generate a pitch track to each interfering source showing the case of one and two speakers.…”