Investigating the effectiveness of speech-to-text recognition applications on learning performance and cognitive load

Yueh Min Huang, Rustam Shadiev, Wu Yuin Hwang

Research output: Contribution to journalArticlepeer-review

31 Scopus citations

Abstract

This study explores the effectiveness of applying speech-to-text recognition (STR) technology during lectures in English on learning performance and the cognitive load of non-native English speaking students. Furthermore, the study also explores the usefulness of texts generated using STR for students with different levels of English as foreign language (EFL) ability during lectures of varying difficulty levels. Two lectures, one with intermediate difficulty level content and the other advanced, were administered, and STR was adopted to aid student learning. The results of this study show that the students who used STR-generated texts outperformed the students who did not. Furthermore, lectures in English caused less cognitive load for low ability EFL students when they used STR-texts. According to the students, the STR-texts were useful for following the instructor, confirming content, clarifying vocabulary, and making up missed information. It was found that STR-texts were used by low EFL ability students during both lectures whereas high EFL ability students used STR-texts during the lecture at the advanced level and only some of high EFL ability students used them during the intermediate lecture. Based on these results, several suggestions and implications for teaching and research community are proposed.

Original languageEnglish
Pages (from-to)15-28
Number of pages14
JournalComputers and Education
Volume101
DOIs
StatePublished - 1 Oct 2016

Keywords

  • Improving classroom teaching
  • Pedagogical issues
  • Post-secondary education

Fingerprint

Dive into the research topics of 'Investigating the effectiveness of speech-to-text recognition applications on learning performance and cognitive load'. Together they form a unique fingerprint.

Cite this