Clinical Linguistics in the Era of Big Data: AI-Assisted Diagnosis of Speech and Language Disorders

Abstract

 Clinical linguistics has traditionally relied on detailed case studies, manual transcription, and expert interpretation to diagnose speech and language disorders. While effective in clinical settings, these approaches are often time-consuming, subjective, and limited in scalability. The rise of Big Data and natural language processing (NLP) has introduced transformative possibilities for clinical linguistics, enabling automated, large-scale analysis of speech samples and linguistic data. This article examines how Big Data-driven AI systems contribute to the diagnosis of communication disorders, with a focus on aphasia, dysarthria, and developmental language disorders. The study utilized three datasets: (1) the AphasiaBank corpus containing transcripts of individuals with aphasia (MacWhinney et al., 2011), (2) a motor speech disorder dataset including dysarthric speech samples, and (3) a developmental language disorder corpus from pediatric clinical assessments. Results demonstrate that Big Data-enhanced models outperform traditional manual methods in diagnostic accuracy and speed, though challenges of data privacy, interpretability, and clinical acceptance remain. The discussion emphasizes that AI-assisted clinical linguistics should complement, rather than replace, expert judgment. By integrating computational models with clinical expertise, Big Data offers significant promise for improving early diagnosis, personalized treatment, and accessibility to speech-language services.