Approximately one-fifth of theworld’s population suffer or have suffered from voice and speech production disorders due to diseases or some other dysfunction. Thus, there is a clear need for objective ways to evaluate the quality of voice and speech as well as its link to vocal fold activity, to evaluate the complex interaction between the larynx and voluntary movements of the articulators (i.e., lips, teeth, tongue, velum, jaw, etc), or to evaluate disfluencies at the language level. The underlying assumption is that deviations from patterns that might be considered normal can be correlated with many different symptoms and psychophysical situations. However, despite a large effort in this field in the last few years, useful services are still far fewer than those in other areas of speech technology (text-to-speech synthesis, speech recognition, speaker recognition and verification, etc.) and, as a result, many results have not been transferred to clinical settings.

中文翻译：

---

基于语音，语音和语言处理的健康状况自动评估问题简介

由于疾病或某些其他功能障碍，全世界约五分之一的人口患有或遭受语音和言语产生障碍。因此，显然需要客观的方法来评估语音和语音的质量及其与声带活动的联系，以评估喉部与咬合架（即嘴唇，牙齿，舌头）的自发运动之间的复杂相互作用。 ，皮肤，下巴等），或在语言水平上评估歧义。基本假设是，与可能被认为是正常的模式的偏差可能与许多不同的症状和心理生理状况相关。但是，尽管最近几年在该领域做出了巨大努力，但有用的服务仍然远远少于语音技术的其他领域（文本到语音合成，语音识别，