| 摘要 |
PROBLEM TO BE SOLVED: To make a document which is converted into a synthesized speech easy to hear when the speech is synthesized from a character string wherein the same symbol characters are described successively. SOLUTION: When a document containing symbol characters is inputted to the preprocessing part 11 of a speech synthesizing device, a segmentation part cuts plural successive symbol character strings out of the character string. The cut symbol character strings are changed into a set processed document 19, or into another reading, a sign tone, a voiceless part, a different speed, different sound quality, and different sound volume by a alteration processing part. Character strings other than symbol characters are processed by a word identification part 13. Then normal processes are performed by a rhythm process part 14 and a speech synthesis part 15 and a synthesized speech is outputted from a speaker 21. Consequently, the symbol character string in the document can be listened to without any feeling of physical disorder or annoyance. When there is one symbol character present independently, its reading is outputted as a synthesized speech. |
