Artificial Intelligence for EFL Speaking Development: A Systematic Review of Empirical Studies (2018–2026)

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Despite the fast expansion of artificial intelligence (AI) in English as a foreign language (EFL) education, evidence regarding its effectiveness in supporting speaking development remains fragmented, with limited synthesis of AI tool types, methodological quality, learning outcomes, and research trends. This systematic review examined empirical studies on AI-supported EFL speaking development and was published between 2018 and 2026. A search of Scopus and Web of Science was conducted using keywords related to AI, EFL, and speaking skills, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Data from 33 studies that met the inclusion criteria were analyzed according to established coding categories aligned with the research questions. The findings indicate that AI tools include conversational agents, speech recognition systems, intelligent tutoring systems, generative AI, and multimodal applications, which each support different aspects of speaking development. Speech recognition tools mostly improved pronunciation, while conversational and generative AI mainly enhanced fluency, confidence, and autonomous speaking practice. Affective gains, including less speaking anxiety and increased willingness to communicate, were more consistent than improvements in more complex speaking skills. This review also found that the evidence base is methodologically inconsistent, concentrated in higher education and specific geographical regions, and increasingly dominated by generative AI technologies.