| 摘要 |
PURPOSE:To estimate the basic accent efficiently on a word composed of two Chinese characters in a Chinese character-Japanese alphabet mixed sentence by estimating an accent pattern as a (1) type when the beat number of a rear element of a two-character idiom is one and as a (0) type when the beat number is two. CONSTITUTION:An accent estimating device 1 is composed, for example, of a computer, and when a two-character word is inputted, reading and an estimated accent of the word are outputted. Here, a character of a rear element (a rear character in two characters) is determined whether it is a single-beat character like 'element' (So), 'character' (Ji), 'part' (Bu) and the like or a two-beat character like 'lighting' (Shout, 'joining' (Goo, 'table' (Hyo) and the like by referring to a character checking table. When the beat number of the rear element of the two-character idiom is one, it is estimated as a (1) type, and when the beat number is two, it is estimated as a (O) type, so that an accent pattern can be estimated accurately in the same thought. |
