Improving EFL learners’ speaking skills and willingness to communicate via artificial intelligence-mediated interactions

It is widely acknowledged that technology, particularly artificial intelligence (AI), presents innovative opportunities for improving English language learning skills. However, little is known about the impact of AI-mediated activities on learners' speaking skills. This study, therefore, aimed to examine the impact of an artificial intelligence platform on English as a foreign language (EFL) learners' speaking skills, including speaking fluency and coherence, lexicon, grammatical range and accuracy, and pronunciation, and willingness to communicate (WTC). EFL learners' attitudes and perceptions towards AI-mediated speaking activities were also qualitatively explored. Thirty-three EFL learners were randomly assigned to an experimental group (AI group) and 32 EFL learners were placed in a control group (face-to-face group). The participants in the AI group engaged in AI-mediated interactive speaking activities using Andy English Chatbot, whilst the face-to-face group engaged in conventional peer-interaction speaking activities. Moreover, IELTS speaking skill tests and a WTC scale were applied to collect the quantitative data followed by an individual semi-structured interview in the qualitative phase. The results indicated that AI-mediated interactive speaking activities were more effective in improving EFL learners' speaking skills and WTC. Moreover, the learners had positive attitudes and perceptions towards the AI-mediated speaking instruction. The study provides valuable insights for language educators and researchers regarding technology-mediated instruction in language classrooms.