Captioned Lecture System for Hearing Impaired Students Using Speech Recognition Technology

摘要

In this research, we developed a captioned lecture system which converts the voice of the speaker in classes, lectures and events at the university into captions in real time and presents them to the hearing-impaired students by means of speech recognition technology and wireless LAN technology. The real-time captions can be viewed on a portable caption display device through wireless LAN regardless of the location. The system was experimentally applied to actual college lectures as a lecture assurance system. With a team of one lecturer and two correctors, caption accuracy was about 93% and the lectures were easy for a hearing-impaired student to understand. Unlike the conventional captioned lecture system using a speech recognition method, there is no need for a specially trained person to repeat the speech, and so the system has higher practicability.