Advancing Speech Disorder Diagnostics: CNN Innovations in Dysarthria Recognition

Neurological disorders such as amyotrophic lateral sclerosis (ALS) and cerebral palsy (CP) are common causes of dysarthria, the subject of this research article. The capacity to communicate is greatly affected by dysarthria, thus early diagnosis is critical for effective therapies and improved patient outcomes. The study employed a Convolutional Neural Network (CNN) for speech classification using the TORGO database, which includes 2000 samples of individuals with and without dysarthria, representing different genders. With 500 samples each, collected over distinct sessions, the dataset contains both dysarthric and non-dysarthric people of both sexes. Distinguishing between dysarthric and non-dysarthric occurrences, the CNN-based approach achieves an astonishing 96% accuracy. This shows how deep learning technology may help with early dysarthria identification and enable rapid therapy for affected individuals. The findings of this study have important implications for the development of diagnostic tools and may lead to better living conditions for those who struggle with speaking.