Abstract:

Nowadays, the number of children suffering from speech disorders and impediments is steadily growing. Difficulties in speech are especially prevalent with children with special educational needs, who might also have limited language skills and may struggle with certain words, formation of phrases and construction of sentences. Traditional speech therapy has seen a lot of innovation, has come a long way in history and is helping people with varying degrees of speech impairments. Technology, however, especially alternative and aided augmentative communication solutions, has succeeded in augmenting therapy and speeding up the improvement of language skills. Many solutions are available in widely spoken languages, such as English, French and Spanish, but for some less common languages, such as Bulgarian, such communication and speech tools are still in a state of development. These languages often exhibit a unique and complex grammar, which, in the instance of Bulgarian, can morph nouns, pronouns and adverbs depending on gender, adjacent singular/plural nouns and membership. This poses unique challenges to the development of grammar-aware software and requires new approaches to the problem. This paper aims to present a knowledge-based algorithm describing the steps required to dynamically form grammatically correct sentences in Bulgarian. This algorithm will be used in a mobile application, which will also provide auditory and graphical cues, making it an especially useful educational tool for children with speech disorders. The main aim of this development is to provide an effective tool for building language skills, especially in young children.

Keywords:

Knowledge-based algorithm, speech disorder, alternative communication, augmentative communication, app.