Development of a speech diagnosis and articulation training system based on the visualization and qualification of dysarthric speech

在本研究中，我们开发了一个基于语言特征可视化和量化的系统来支持困难运动障碍的诊断和语言训练。在可视化方面，开发的基于pc的工具能够使用我们的语音可视化技术对音素失真进行可视化表示和量化。此外，我们还设计了一个元音发音训练的原型工具，使用合成元音的目标音来模拟自己的发音。在诊断功能方面，我们将在口腔外科领域开展临床评价的实验，开发的元音训练工具也已在一些聋哑学校得到应用。

In this study, we developed a system to support the diagnosis and the speech training for the dysarthric disorders based on visualization and quantification of the speech features. In visualization, the developed PC-based tool is capable of visual representation and its quantification of phoneme distortion using our speech visualization techniques. In addition, we have designed a prototype tool for vowel articulation training using a target sound of the synthetic vowel to simulate oneself utterance. In diagnostic function, we will start the experiment in the clinical evaluation at oral surgery field, and also the developed vowel training tools have been used in some deaf schools.