In this paper we describe an improved version of our previously proposed feed-back Residual Interfering Sign4 Canceler (RISC) algorithm that can now track the phases and amplitudes of the sinewave components present in a signal without any a priori knowledge. We also give the details of the feed-forward RISC algorithm for tracking the sinewave components prt,sent in voiced speech. The effectiveness of these algorithms is demonstrated by processing both synthetic and clean natural speech. © 1994 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.