Author: Sharipova, Aziza Abdumanapovna; Abdullayeva, Malika Ilkhamovna

Annotation: Children with speech impairments often face significant challenges in acquiring foreign language skills, particularly English, due to limitations in phonemic awareness, articulation and auditory processing. Recent advances in artificial intelligence (AI) and machine learning enable the creation of adaptive, individualized rehabilitation tools that support both speech therapy and language learning. This study shows the development and initial evaluation of an AI-based computer program designed to improve English pronunciation, vocabulary acquisition and speech clarity among children aged 6–12 with diagnosed speech impairments. The system integrates automatic speech recognition (ASR) optimized for atypical speech, a real-time pronunciation feedback module, gamified learning tasks and therapist-controlled customization. Preliminary results from a pilot study (12 children were involved) indicate improvements in articulation accuracy, motivation, and English phoneme recognition. The findings highlight the potential of AI-driven educational technologies to complement traditional speech therapy.

Keywords: artificial intelligence, rehabilitation, ASR, speech impairments, Adaptive Learning, Gamified Learning, pronunciation, phoneme recognition, inclusive education.

Pages in journal: 509 - 517