Speaker
Description
In recent years, AI-powered applications in speaking instructions have been growing in popularity due to their interactive features which help engage learners and encourage regular practice. Therefore, many educators have integrated AI tools into their teaching to facilitate instructions and improve learning outcomes. This study investigates the impact of SpeakEasy AI, an AI-powered web application that enables English learners to practice speaking through multiple modes including real-time conversation, simulated tests, and speech audio assessment. A mixed-method experimental design was employed, involving 80 first-year English-major students at Ho Chi Minh City University of Foreign Languages - Information Technology (HUFLIT). The experimental group used SpeakEasy for 10 weeks, while the control group received traditional instructions. Quantitative data were collected through pre-tests, post-tests and questionnaires, and the effectiveness of SpeakEasy AI was compared with conventional teaching methods using statistical analysis. In addition, in-depth interviews were conducted to obtain qualitative insights into students’ experiences and perceptions. The results revealed that the experimental group exhibited significantly greater improvement in fluency and overall speaking competency compared to the control group. Additionally, the experimental group reported positive perceptions of the application’s effectiveness, particularly in reducing anxiety and enhancing speaking confidence. The interview data also indicated additional benefits beyond the enhancement in speaking performance, along with some recommendations to optimize the application. Further research is required to examine its long-term effects across more diverse populations of EFL learners.
Key words: SpeakEasy AI app, AI-powered chatbot, speaking fluency, speaking competency, EFL learners.
