Apparatus speech synthesis (10) for synthesizing speech consisting of a plurality of speech segments, each of which comprises at least one phoneme, said apparatus speech synthesis (10) characterized by comprising: base data (111) for storing data related to said speech segments; decomposition means speech segment (101) of disassembling each of said speech segments into a plurality of tone signals, each of which has a phase characteristic; means for generating phase characteristics (102) for generating a uniformed phase characteristic by averaging said phase characteristics of said pitch waveforms obtained by said decomposition means speech segment (101); means for transforming the phase characteristics (103) for transforming said phase characteristics of said pitch waveforms into said uniformed phase characteristic generated by dichosmedios generation phase characteristics (102); sorting means tone signals (104) for classifying said pitch waveforms into a plurality of groups; recording means tone signals (106) for registering said pitch waveforms in said database (111), extracting one pitch waveform from among said pitch waveforms in each of said groups; and synthesizing means (107) for synthesizing said speech with said pitch waveforms registered in said database (111).
展开▼
