摘要 |
PROBLEM TO BE SOLVED: To generate a compound voice of unclear meaning while holding language likeness in view points of a rule of composition, a phoneme and a rhythm. SOLUTION: An input text is analyzed referring to a word dictionary constituted of a set of a character sequence expressing a word, a phoneme sequence expressing reading of the word, and a part of speech of the word, the input text is divided into partial texts, a phoneme sequence and a part-of-speech sequence are generated in every partial text, the part-of-speech sequence of every partial text is collated with a list of the part-of-speech sequences, so as to determine whether converting the phoneme of the partial text or not, and each phoneme is converted in the phoneme sequence of the partial text determined to convert the phoneme. COPYRIGHT: (C)2009,JPO&INPIT
|