Method and device for the synthesis of the speech by adding - overlap of the wave forms

摘要

Speech synthesis is carried out on the basis of elements such as the diphones. It is carried out, at least on the voiced sounds of the sound elements, an analysis by windowing substantially centered on the start of each impulse response of the vocal tract to the excitation of the vocal chords with the aid of a filtering window having an amplitude decreases down to zero at the edges of the window of width at least equal to twice the fundamental period of origin or twice the fundamental period of the synthesis, it replaces the signals resulting from the "windowing " corresponding to each of the sound element, with a time offset from those - ci is equal to the fundamental period of the synthesis, less than or greater than the fundamental period of origin, according to the prosodic information relating to the frequency of the synthesis is carried out the synthesis by summing the signals thus offset.