Pronunciation Tutor for Deaf Children based on ASR

Abstract

ASR, whose full name is Automated Speech Recognition, is a technology that converts human speech into text. Speech recognition, a multidisciplinary field, is closely related to acoustics, phonetics, linguistics, digital signal processing theory, information theory, computer science and other disciplines. ASR has been applied in educational technology such as deaf children's education in this day and age. This paper makes a preview of a project in which a computer-aided tutor for deaf children instruction based on the speech recognition technology. This tutor utilizes three effective models and is combined with data mining technology. Two evaluation approaches and overview of embedded experiment are also detailed in this paper.