基于舌部可视化模型的听障儿童发音训练

Articulation training for children with hearing disorder based on visualization model of tongue

基于可视化模型提出了一种听力障碍儿童的舌部训练方法。通过对听障儿童舌部发音位置的训练,辅助其获取正确的发音位置和发音方式。该方法采用三维建模技术和传感器技术相结合的舌部康复训练方法,建立舌部三维模型;用腭位传感器获取发音过程中舌部位置信息;选取23个声母作为样本,建立健听儿童声母发音舌位置信息样本数据库;最后,驱动舌部模型,引导听障儿童进行舌部发音位置训练。该方法能够显示健听儿童发音时舌部的运动过程,克服了目前听障儿童语言训练中舌部训练过程中不可视的问题。实验结果表明:该方法提高了听障儿童发音训练的效果,缩短了康复训练时间,尤其对于舌、下腭之间配合较为复杂,发音位置较为隐蔽的发音,提高发音训练的准确度超过15%。

An articulation training method for children with hearing disorder was proposed based on a visualization tongue model.By training on the position of articulation for children who have residual hearing and hearing compensation,these children could acquire the accurate place and way of articulation.A tongue three dimensional model was established based on the tongue rehabilitation trainingwith the combination of three-dimensional modeling technology and sensor technology.The three-dimensional positions of tongue during the process of articulation was acquired by apalate sensor.Then,23 initial consonants were selected as samples for establishing a sample database of tongue positions of children with normal hearing during articulation.Such the tongue model could be activated to provide a guidance for children with hearing disorder during training on the position of articulation.This method displays the movement of tongues of children with normal hearing during articulation and overcomes the shortcomings that has been neglected in tongue training and language training.Experimental result indicates that this method improves the effect of articulation training for children with hearing disorder and shortens training time.Especially for those articulation processes involving complicated coordination between tongue,lower jaw and unobvious position of articulation,the method increases the accuracy of articulation training by 15%.