基于生理发音模型的中文发音可视化技术研究-计算机科学与技术专业论文.docx

ABSTRACT Pronunciation process visualization refers to displaying the movement of the articulators during pronunciation process vividly. Pronunciation process visualization can be applied in many fields, for example in the field of language learning. Spreading Chinese language worldwide is not only in line with national strategy, but also an important way to strengthen the national “soft power”. NOCFL (China National Office for Teaching Chinese as a Foreign Languge) has established a number of Confucius Institutes in the world. Yet it is still difficult to meet the needs of more than 40 million overseas Chinese language learners. A great deal of Chinese language teachers is still needed. In addition, in the field of speech therapy, there are tens of millions of people with speech disorders. The technology of pronounciation process visualization can help the above patients correct their pronouciations better through visual feedback. For the above background, this paper presents a method of using Chinese electromagnetic Articulographic (EMA) data to drive three-dimensional (3D) physiological model to achieve the method of Chinese pronunciation process visualization. To begin with, the data were collected. On the one hand, EMA database covering 30 Chinese phonemes was established, and then a software for data annotation was developed to label EMA database in phoneme level. The acoustic parameters and articulatory parameters were carried out to refine data. On the other hand, for the 3D and physiological pronunciation model, nearly 150,000 muscle combination files were designed in order to cover the entire pronunciation space. These files were input to the 3D physiological model and then the corresponding articulatory data were gained. Secondly, before mapping from the EMA data to 3D model data, the Thin-plate Spline method was used for adjusting the coordinate system. The EMA data of Chinese neutral vowel [?] and the data of 3D model at its resting state were compare