摘要 |
PROBLEM TO BE SOLVED: To provide a speech synthesizer and a program thereof capable of creating synthesis sound even when including no digraph matching with an input sentence.SOLUTION: A speech synthesizer 1 is a device for synthesizing a voice corresponding to an input sentence using a digraph which is constituted of plural nodes and edges each representing a connection between nodes and is capable of forming a sentence using text data each of which can be replaced with the respective nodes and plural voices corresponding to the text data. The speech synthesizer 1 includes: a graph group storing unit 10; an input sentence graph comparison unit 20; a graph addition unit 30; an input sentence addition graph comparison unit 40; a voice selection unit 50; and a voice connection unit 60. |