A superpositional model applied to F0 parameterization using DCT for text-to-speech synthesis | IEEE Conference Publication | IEEE Xplore