This work presents the development of an ontology for speech disorders in children, in order to become a tool to support therapists for diagnosis and possible treatment. Speech disorders are classified using a taxonomy obtained from a speech disorders corpus previously conformed. Based on this taxonomy, the ontology, which structures and formalizes concepts defined by the main topic authors, is developed. The ontology's main classes represent the taxonomic classification of speech disorders, their etiological origin, symptoms, and signs of each disorder, assessment, and intervention strategies; it also represents patients as it instances. A transcription module is also used to make different pronunciation tests and to obtain more detail of the characteristics presented by each patient to make the diagnosis. The development of the tool and the transcription module is based on Natural Language Processing (NLP) and Information Retrieval (IR) techniques. The importance of an early detection and diagnosis of a speech disorder-which can have a social, economic and educational impact-, lies in the fact that the prognosis of the treatment depends on the cause of the disorder and on an opportune treatment.