PURPOSE: To display and output a synthesized voice synchronizing with the motion of a mouth-figure image by comprising a system so as to switch the mouth-figure image setting the pronouncing time of each syllable of a regular voice synthesis part when the syllable is inputted at the outside of the regular voice synthesis part.
CONSTITUTION: Sentence data disassembled at a sentence disassembling part 1 is supplied to the regular voice synthesis part 2 comprehensively, and synthesized voice output is generated by regular voice synthesis processing, however, when such unified sentence data is sent to a pronouncing time memory part 6, the pronouncing time memory part 6 outputs the pronouncing time of each syllable of the regular voice synthesis part 2, and supplies it to an image display control part 4. Thereby, the image display control part 4 outputs a prescribed mouth-figure image from image memory 5, however. since the pronouncing time memory part 6 stores the pronouncing time to be used of each syllable of the regular voice synthesis part 2 in advance, the pronouncing time represents the paragraphic time of the syllable, therefore, a natural mouth-figure image can be outputted without taking detailed handshake between the image display control part 4 and the regular voice synthesis part 2.
NAKAGAWA AKIRA
MATSUDA KIICHI
