Synthesizing intonation using the RFC model.
Taylor, Paul A
This paper describes an implementation of the rise/fall/connection (RFC) model of intonation for use in the speech synthesis component of a machine interpretation system. The phonological input is described, as is the algorithm which converts this to a rise/fall/connection description, and eventually to an F0 contour. As we wish to model different speakers’ intonational characteristics, some details are described on the method for adapting the model’s parameters to match the behaviour of a given speaker.