Exploring technology acceptance of ASR for pronunciation learning

Abstract

Improving pronunciation in second language (L2) learning requires timely and personalized feedback but educators have limited time to dedicate in the classroom. While there are many educational tools available, educators are still struggling to integrate technology in their practices. To address this challenge, this study explores repurposing a freely available ASR tool for pronunciation learning. Despite research demonstrating ASR's effectiveness, factors influencing educator and student decisions to accept these tools, remain unclear. Framed by the Technology Acceptance Model (TAM) and the Technological Pedagogical Content Knowledge framework (TPACK), this study uses mixed methods to investigate these factors. Data from multiple sources, such as speech samples, interviews, surveys, screen recordings, and observation notes, were analyzed using statistical analysis and coding frameworks. Key findings showed that educators’ acceptance is influenced by several factors, such as, accessibility, knowledge of technology, pedagogy, and pronunciation, and motivation. The findings regarding the students’ experiences suggest a strong link between technology acceptance and pronunciation improvement, while the relationships with TPACK and students’ interaction with the tool were moderate to weak. Additional factors were identified through analyzing the screen recordings and the interviews, such as, personality traits, personal beliefs and values, and emotions evoked by the experience. The study argues for a modified TAM model, called STAM, which incorporates both qualitative and quantitative data to systematically explore technology acceptance. This model can guide future research and the integration of repurposed technologies in the L2 classroom. Moreover, this work also informs the design of future ASR tools to better support pronunciation teaching and learning, facilitating broader advancements in computer-assisted pronunciation teaching.