This study aimed to investigated whether speaking practice with text-to-speech (TTS) technology would decrease students’ anxiety and increase their confidence associated with making presentations in English in front of their classmates in a university EFL course. Previous studies have demonstrated the usefulness of TTS in language learning settings. For example, Moon (2020) conducted a questionnaire survey on self-generated listening materials based on TTS, and found that most students reported that TTS-based listening materials reduced their listening anxiety and increased their confidence. Although previous studies have investigated the usefulness of TTS in language-learning settings, no study has compared different models of TTS-synthesized speech and their effects on the emotional aspects of learner engagement. Therefore, in this study, two TTS-synthesized speech models were developed. The first was based on the utterances of a native English speaker, and the second on utterances from a non-native Japanese English speaker. In addition, we developed an interface that allowed learners to use these two models. Two classes of first-year students (upper: 41, lower: 35) from a Japanese university voluntarily participated in this study. Students were encouraged to practice making their presentations in and out of class according to the following three steps: (1) check the pronunciation of the English in the manuscript with the TTS-synthesized speech, (2) read the manuscript aloud while listening to the TTS-synthesized speech, and (3) speak while listening to the TTS-synthesized speech. In late July 2023, a post-questionnaire survey was administered regarding TTS use for speaking practice, students’ anxiety associated with it, and students’ confidence in making presentations in English. The results revealed that both upper and lower student groups rated TTS-synthesized speech as useful for presentation practice, and both expressed the view that using TTS-synthesized speech to practice their presentations reduced their anxiety. The lower group strongly believed that practicing their presentations using TTS-synthesized speech made them feel less concerned about their own English pronunciation, and they attached more than twice the value to TTS-synthesized speech by native English speakers than they did to Japanese English speakers.

---

Abstract Summary
This study investigated whether speaking practice with text-to-speech (TTS) technology would decrease students’ anxiety and increase their confidence associated with making presentations in English in front of their classmates in a university EFL course. In this study, two TTS-synthesized speech models were developed. The first was based on the utterances of a native English speaker, while the second was based on utterances from a non-native Japanese English speaker. Two classes of first-year students (upper and lower) participated voluntarily in the study. The results of a post-questionnaire survey revealed that both the upper and lower groups reported that using TTS to practice their presentations reduced their anxiety, and the lower group strongly felt that practicing their presentations using TTS made them feel less concerned about their own English pronunciation. They valued the TTS-synthesized speech of native English speakers more than twice as much as that of Japanese English speakers.

Authors:
Takatoyo Umemoto, Kyoto University of Foreign Studies, Japan
Shinnosuke Takamichi, The University of Tokyo, Japan
Yuta Matsunaga, The University of Tokyo, Japan
Yusuke Yoshikawa, Kyoto University of Foreign Studies, Japan
Shigeo Fujiwara, Uchida Yoko Co., Ltd., Japan
Yasushige Ishikawa, Kyoto University of Foreign Studies, Japan

About the Presenter(s)
Dr Yasushige Ishikawa is a Professor at Kyoto University of Foreign Studies, Japan. His research interests include CALL (Computer-Assisted Language Learning), MALL (Mobile-Assisted Language Learning) and corpus linguistics.

See this presentation on the full schedule – Sunday Schedule

Bookmark