The efficacy of linear prediction for speech analysis depends upon the degree to which the speech signal is stationary within an analysis interval. For this reason the speech signal is usually divided into small intervals during each of which the signal is assumed to be stationary. The formulation described in this paper is aimed at accomodating the nonstationarity more automatically by allowing the linear estimate of the speech signal to be time varying. Subject Classification: 70.40, 70.65.
No takes yet. Share an insight, caveat, or question.
Louis A. Liporace (1975) studied this question.