发明名称 SPEECH SYNTHESIS DEVICE, SPEECH SYNTHESIS METHOD AND COMPUTER PROGRAM
摘要 PROBLEM TO BE SOLVED: To provide: a speech synthesis device for generating synthesized speech from text data; a speech synthesis method; and a computer program. SOLUTION: Text data collected by a text collection section 163 are divided into morphemes by a morpheme analysis section 164. A frequency calculation section 165 counts the number of occurrences of the morpheme including each Chinese character stored in a detailed reading dictionary 2b, in the morphemes divided by the morpheme analysis section 164. A word selection section 166 selects the morpheme having the largest number of occurrences calculated for the each Chinese character stored in the detailed reading dictionary 2b, and a detailed reading information generation section 167 generates detailed reading information for each Chinese character based on the selected morpheme. A dictionary update section 168 updates the detailed reading dictionary 2b based on the detailed reading information generated. The speech synthesis device achieves the detailed reading by imparting the detailed reading information stored in the detailed reading dictionary 2b to a predetermined Chinese character when generating synthesized speech from the text data. COPYRIGHT: (C)2010,JPO&INPIT
申请公布号 JP2010181562(A) 申请公布日期 2010.08.19
申请号 JP20090023919 申请日期 2009.02.04
申请人 FUJITSU LTD 发明人 KOJIMA HIDEKI
分类号 G10L13/08 主分类号 G10L13/08
代理机构 代理人
主权项
地址