Personalized learning pathways in AI-powered dubbing applications for speaking proficiency enhancement

The integration of artificial intelligence in language education has revolutionized pedagogical approaches, with AI-powered dubbing applications emerging as promising tools for developing speaking proficiency through personalized learning pathways. This systematic review synthesized evidence from 38 empirical studies involving 4,327 participants to evaluate the effectiveness of personalized learning pathways within AI-powered dubbing applications for Business English speaking proficiency enhancement. Following PRISMA guidelines, comprehensive searches across seven databases identified peer-reviewed studies published between 2019-2024, with quality assessment employing Cochrane risk-of-bias tools and meta-analysis conducted where appropriate. The analysis revealed substantial improvements in pronunciation accuracy (Cohen’s d=1.82, 95% CI: 1.65-1.99) and fluency development (d=1.46, 95% CI: 1.29-1.63), with intermediate-level learners demonstrating 68.4% greater gains compared to advanced learners. Subgroup meta-analysis confirmed neural network superiority over collaborative filtering approaches, achieving 87.3% accuracy in pronunciation feedback. Publication bias assessment revealed asymmetrical distribution (p=0.031), though trim-and-fill analysis indicated minimal impact on primary conclusions. Cost-effectiveness analyses demonstrated significant advantages, requiring $15-25 per student annually compared to $180-240 for equivalent individual tutoring. Cultural engagement patterns aligned with Hofstede’s dimensions theory, where East Asian learners showed higher completion rates but lower self-efficacy scores. Despite documented learning plateau effects after 4-6 weeks, AI-powered dubbing applications demonstrate significant potential for enhancing speaking proficiency, though optimal implementation requires hybrid approaches integrating human pedagogical expertise with technological affordances to address cultural contextualization and sustained engagement challenges.