This paper presents the AmaLgAM App, an innovative digital tool developed within the AmaLgAM research project, designed to assess the narrative skills of preschool children with Autism Spectrum Disorder (ASD). The application utilizes Artificial Intelligence (AI) to automatically transcribe, linguistically analyze, and conceptually evaluate children’s oral responses during storytelling and comprehension activities. Narrative ability is a key indicator of linguistic, cognitive, and social development. However, current assessment practices are often manual, time-consuming, and subject to evaluator bias. AmaLgAM App addresses these limitations by offering an automated, objective, and scalable solution. The app engages children in a story-based interactive environment and records their verbal responses, which are then analyzed for linguistic complexity, coherence, and emotional insight. Developed through a participatory co-design process (Computational Linguistic Living Lab – CLLL), the application incorporates input from educators, therapists, parents, and autistic individuals to ensure accessibility, relevance, and usability. The tool is tailored to the needs of professionals in education and therapy, enabling detailed tracking of each child’s progress over time.

The AmaLgAM App combines pedagogical and technological innovation, supporting early language assessment and intervention in naturalistic, child-friendly settings. It represents a significant step toward automated, ecologically valid evaluation of language abilities in neurodiverse populations.

Keywords: Νarrative Skills, Artificial Intelligence, Autism Spectrum Disorder, Language Assessment, Preschool, AmaLgAM App.