Disruption of speech in children of preschool age has become a serious problem of society. The number of parents visiting special centers and specialists is increasing year by year. To solve this problem, we can develop new technologies in the Kazakh language using natural language processing and machine learning. Connecting to the information flow of this area and working with new technologies will obviously contribute to the achievement of effectiveness in the development of the child. That is why the importance of creating an information system in the field of speech correction increases.

The article describes the method of building a system of teaching the Kazakh language to children with impaired speech abilities, including a dictionary of synonyms of the Kazakh language. The dictionary of synonyms works on the development of speech skills correctly and in the native language, increasing vocabulary depending on the level of the child. The novelty of the proposed approach lies in the identification of semantically similar words in texts in the Kazakh language. This work helps to automate the new pedagogical system in the field of speech therapy.