Method and apparatus for speech synthesis and program recorded medium

Abstract

Data in the same range of the fundamental frequency F.sub.0 as speech segments are used as learning data to prepare a reference codebook CB.sub.M for a spectrum envelope. The same learning data for a higher range than F.sub.0 and the same learning data for a lower range are subject to a linear stretch matching with respect to the learning data for the range F.sub.0. For each vector code in the reference codebook CB.sub.M, the spectrum envelope is clustered to prepare a high range codebook CB.sub.H and a low range codebook CB.sub.L. The spectrum envelope of input speech segments are fuzzy vector quantized (S402) with the reference codebook, and depending on the synthesized F.sub.0, a high, middle or low codebooks is selected. The selected codebook is used to decode the fuzzy vector quantized code, and the decoded output is subject to the inverse FFT. Alternatively, codebooks CM.sub.MH and CB.sub.ML each comprising differential vectors for corresponding code vectors between CB.sub.M and CB.sub.H and between CB.sub.M and CB.sub.L are prepared. The quantized code is decoded using either CB.sub.MH or CB.sub.ML, and the decoded differential vector is stretched in accordance with a difference in the fundamental frequency between the synthesized speech and the original speech for CB.sub.M. The stretched differential vector is added to the code vector which was used for the fuzzy vector quantization.