Effects of text-to-speech synthesized speech on learners’ presentation anxiety and self-efficacy: A comparison of two models

Abstract

[EN] This paper reports on the effects of two Text-To-Speech (TTS) synthesized speech models, one based on English utterances by a native English speaker and the other based on English utterances by a Japanese non-native speaker of English, on presentation anxiety and self-efficacy of Japanese English as a Foreign Language learners. We hypothesized that learners’ presentation anxiety would decrease and their self-efficacy would increase when using the Japanese non-native English speaker model compared with the native English speaker model. 55 first-year university students (upper level: 33; lower level: 22) voluntarily participated in the study and were divided into experimental and control groups. To measure the participants’ presentation anxiety and self-efficacy, the scale developed by Ishikawa et al. (2021) was used. A mixed-design three-factor ANOVA with the group, class level, and period as independent variables showed an interaction between class level and period. A simple main effect test indicated a significant increase in self-efficacy in the upper-level students. These results reveal that, regardless of the model used, the use of TTS-synthesized speech significantly increases the self-efficacy of the upper-level students. The paper concluded that further research on technology use and learner affect needs to be conducted.