A Simple Method of Unvoiced/Voiced Classification of Speech Signal
Ying Guo · Journal of Air Force Engineering University · 2002
This paper presents a simple method of unvoiced/voiced classification based on combinatorial parameters, which decreases the overlap area of the characteristic probability density of the unvoiced/voiced speech signals. Simulations on computer show that the method is more efficient and robust than that of classification based on single parameter, and the design of the program is simple and concise.