Pinyin Automatically Marked For Multi-tone Words In Modern Chinese Study

In modern Chinese, the modern Chinese words including multi-tone components are often mispronounced. They will affect people's learning and work. Because of called"multi-tone", it is in essence to the words as written in terms of units. It corresponds to the language unit may be a free morpheme, free morpheme, or may be non-syllable non-morpheme. To facilitate the article, it is said the"more modern Chinese sounds the word"to refer to multi-tone components concerned in modern Chinese words. In modern society, the application of computers and the Internet have been around almost every corner. Softwares of words in Modern Chinese pinyin tone mark have emerged. For the special nature of multi-tone of mark words, the current pinyin software automatically takes from different aspects of measures to make the words sound more different degrees of phonetic issues resolved. It also offers people the fast-paced life a convenient way of learning. However, for various reasons it can not be completely phonetic annotation software and sound of words in modern Chinese pronunciation more correct identification of the existence of such kinds of problems."Modern Chinese Dictionary"(fifth edition) in more than nine hundred and referencing to"form of modern Chinese words,"are polyphone-based. Then, they are collected in Contemporary Chinese Corpus at Peking University, and draw a certain number of corpuses in the article. On this basis, more than 50 high-frequency sound words'corpus are selected and they are given out the Pinyin spelling software measurements one by one. Afterwards, the results are described and analyzed.The article is divided into three parts. Above all, the multi-tone polyphonic characters and words are defined as these two concepts. Next, the number of multi-tone character and further classification of multi-tone word is determined. Secondly, the more concerned tone in the Pinyin spelling words in the corpus of the test software conduct the amount of statistical investigation. The correct pronunciation of the different label and the number of errors marked is specifically listed in the corpus. Finally, the corpus with specific problems one by one is observed, which characteristics are described and rule is summarized.