Application of large language models in the field of children’s language disorders
大语言模型在儿童语言障碍领域的应用

摘要

儿童语言障碍主要包括发展性语言障碍（DLD）、特殊型语言障碍(SLI)、自闭症谱系障碍（ASD）相关的语言障碍等类型。语言障碍是儿童语言发展中的常见问题，具体表现为语言理解或表达能力的显著落后，交流过程中出现障碍。根据世界卫生组织统计，语言发育迟缓儿童发生率约为6—8%。我国语言障碍儿童占儿童总数的比例估计达到10%-20%。儿童语言障碍已经成为一个急需解决的问题。传统干预方法依赖人工评估与一对一训练，效率低且难以实现个性化支持，导致治疗效果不理想。近年来，以DeepSeek-R1为代表的大语言模型技术的快速发展为语言障碍儿童的诊断、干预及教育支持提供了全新路径。本文结合文献研究与应用案例，探讨大语言模型在语言障碍领域的应用现状、技术优势及未来挑战，旨在更好地服务语言障碍儿童。

Abstract

Children’s language disorders mainly include developmental language disorders (DLD), specific language disorders (SLI), and language disorders related to autism spectrum disorder (ASD), etc. Language disorders are a common problem in children’s language development, manifested as a significant lag in language comprehension or expression abilities, as well as obstacles in communication processes. According to the World Health Organization, about 6% to 8% of children worldwide have language disorders. It is estimated that the proportion of children with language disorders in China accounts for 10% to 20% of the total number of children. Children’s language disorders have become an urgent issue that needs to be addressed. Traditional intervention methods rely on manual assessment and one-on-one training, which are inefficient and difficult to achieve personalized support, resulting in unsatisfactory treatment outcomes. In recent years, the rapid development of large language modeling technology represented by DeepSeek-R1 has provided a new path for the diagnosis, intervention, and educational support of children with language disorders. This article combines literature review and application cases to explore the current application, technological advantages, and future challenges of large language models in the field of language disorders, aiming to better serve children with language disorders.