A method of fitting and synthesizing speech intonation by means of Legendre orthogonal polynomials has been proposed and simulated. When applied to pitch curves and intensity curves, the fitting error was small. Using the fitting coefficients, the in tonation of different types of spoken Chinese sentences was syn thesized satisfactorily.
Johnson, L.W. and Riess, R.D.Numerical AnalysisAddison-Wesley Publishing Company , Reading, Massachusetts (1977), 221-223.
2.
Market, J.D."The Sift algorithm for fundamental frequency estimation."IEEE Trans on Acoustics, Speech and Signal Processing. AU-20 (December 1972), 367-377.
3.
Sampei, J. and Asada, A."High quality Parcor speech synthesizer."IEEE Trans Consumer Electronics. CE-26 (August 1980), 353-359.
4.
SIMON, J C. (ed.) Spoken Language Generation and Understanding.D. Reidal Publishing Co., Dordrecht, Holland (1980).
5.
Suen, C.Y.Computational Analysis of Mandarin, Birkhauser Boston (1979).
