Parameter distributions for speech signals modeled with spherically invariant random processes
Phillip L. De Leon, Hui Lan Jiang · 2003
Spherically Invariant Random Processes (SIRPs) have been used as a statistical model for speech and shown to better fit measured probability density functions. With the G-function, SIRPs are conveniently represented and applications involving speech (SIRPs) can be analyzed. This paper describes work in characterizing G-function SIRP parameter distributions for American English.