Integrating an ASR-based translator into individualized L2 vocabulary learning for young children

Vocabulary learning is important to learners of a second language (L2). Nonetheless, organizing young children to practice L2 words in a classroom setting might cause many challenges to educators and practitioners. New technologies such as Automatic Speech Recognition (ASR) may be utilized to solve the issue. In this study, we present a pedagogical innovation of using an ASR-based translator to assist young children to practice L2 words in a personalized and engaging way. ASR-based translators, in general, are easy to access in various smart devices such as smart phones, tablets, intelligent personal assistants and social robots. Although they are not designed to be educational tools, they might be adapted creatively into pedagogical practices. In the current study, we evaluate the effectiveness of word training with an ASR-based translator. Twenty-four Australian young children with an average age of 8.21 years old were assigned into two conditions: (1) the experimental condition used the ASR-based translator (group E), and (2) the comparison condition adopted a traditional teacher-led way of training (group C). Both quantitative data and qualitative data were collected and analyzed. It was discovered that group E significantly outperformed group C in the immediate post-tests, while the superiority became less obvious in the delayed post-tests. In addition, the young children in group E enjoyed training with the ASR-based translator. Implications for future practice and future research are discussed.