PROBLEM TO BE SOLVED: To add speech tone expressing a speaker's mental attitude, in a voice editing composite device for combining recorded voice and rule composite voice to create voice data.;SOLUTION: A voice editing composite system stores first voice data which is voice data of the first text issued with a first speech tone added, stores information showing a first acoustic feature amount which is the acoustic feature amount of the first voice data, and stores edit information showing the variation amount between the first acoustic feature amount and a second acoustic feature amount as the acoustic feature amount of second voice data created by adding a second speech tone to the first voice data. The voice editing composite system creates a third acoustic feature amount as the acoustic feature amount of the third voice data added with the first speech tone by applying rhythm model to a second text, creates a fourth acoustic feature amount by superimposing the edit information on the third acoustic feature amount, and outputs voice to the output device based on the fourth voice data having the fourth acoustic feature amount.;COPYRIGHT: (C)2009,JPO&INPIT
展开▼