论文部分内容阅读
本文介绍了一种新型的包含虚拟发音人头像的语言训练系统并对该系统在听障儿童语言康复训练中的实用性进行了研究。该系统以虚拟三维发音人头像的形式再现了汉语中常见的20个单音节字的发音动作,并将之用于语言康复训练。系统运用EMA(三维电磁发音仪)设备采集内部及外部发音器官运动轨迹作为驱动数据,有效模拟了汉语发音。此外,系统还包括一个听力训练模块和一个朗读训练模块,用于辅助聋儿的发音康复训练。文章对系统发音模拟功能的准确性进行了人工评估,并应用系统对两名4至5岁之间的聋儿进行了初步的康复训练。初步评估及训练的结果表明:系统的发音模拟功能可有效模拟发音器官的动作,有助于提高听障儿童的语言能力;同时,系统的设计风格较易获得幼儿认同。研究证明,本系统在听障儿童发音训练方面是有效的,具有较高的应用价值。
This paper introduces a new type of language training system that includes the avatars avatars and studies the practicability of this system in language rehabilitation training for hearing impaired children. The system reproduces the pronunciation of 20 common monosyllabic words in Chinese in the form of virtual three-dimensional pronunciation human heads and uses it in language rehabilitation training. The system uses the EMA (three-dimensional electromagnetic generator) device to capture the movement trajectories of internal and external vocal organs as driving data, effectively simulating Chinese pronunciation. In addition, the system includes a listening training module and an aloud training module to assist deaf children in pronunciation rehabilitation training. In this paper, the accuracy of systematic phonetic simulation was evaluated manually. The system was used to conduct initial rehabilitation training for two deaf children aged 4 to 5 years old. Preliminary assessment and training results show that: the system pronunciation simulation function can effectively simulate the movement of pronunciation organs, help improve the language ability of hearing-impaired children; at the same time, the system design style is easier to get children recognition. Studies have shown that this system is effective in pronunciation training of hearing impaired children and has high application value.