Computer Discriminative Acoustic Tool for Reading Enhancement and Diagnostic: Development and Pilot Test

Abstract

Speech is a simple and effective method of communication between humans, currently we are not only connected physically but also to the many devices in our life. Designing a machine that mimics human capabilities, especially in listening and responding to uttering words of humans have puzzled engineers and scientist for decades. In this paper, a thorough analysis of the literature on the acoustic was conducted, and the general architecture of the automatic speech recognition is presented consisting of acoustic front-end, acoustic model, language model, and decoder. In the test implementation of the speech recognition system, the Technology Acceptance Model was utilized, and Post-Study System Usability Questionnaire (PSSUQ) was used in gathering the feedback of the initial users. Further, experimentation to selected students was administered in a four-week setting and divided into two groups controlled and uncontrolled. The result is encouraging in the two measures: acceptability and implementation, further study is encouraged in a broader sample is sought. This will be helpful in teaching reading especially in this time of pandemic where face-to-face is limited.