Using Windows Speech Recognition Tool to Improve English Speaking Skills of Undergraduates Learning English as a Second Language

Abstract

Undergraduates who pursue their main stream studies in English find the ability to express themselves in English imperative not only for day-to-day communication in the academic setting but also as a prerequisite skill as future professionals. However, many students complain of being extremely nervous and apprehensive when assigned speaking tasks and consequently switching back to express themselves in the first language. The key objective of this empirical research is to test the usability of Windows Speech Recognition Tool as a resource to improve English speaking skills of undergraduates in an ESL context. The Windows Speech Recognition Tool is a program originally designed by Microsoft to command personal computers with the users‘ voice and to dictate voice to word documents. It was hypothesized that the practice gained by learners when interacting with the computer in English speaking tasks would pave the way to communicate with peers and mentors in English more confidently. A sample of 30 undergraduates who had scored comparatively low marks for the speech component in an English language test was selected. An induction was given to the students about the use of the speech recognition tool, for instance, the correct use of microphone and the need to pronounce words correctly in order to be recognized by the tool, etc. Subsequently, each student was assigned an array of tasks to be completed at their own pace. These assignments consist of activities that require the use of speech recognition tool in order to comment on videos, describe given pictures, and to give instructions to carry out medical tests, etc. The scripts of the speech that the students produce are recorded by the tool and the students are given feedback for improvement accordingly. The students face a speech evaluation test conducted by the mentor at the end of the course. These marks will be comparatively analyzed using the t-test strategy with the marks of the pre-test.