AI-Powered MOOCs: Video Lecture Generation

Abstract

This paper focuses on the application of artificial intelligence in education that artificial intelligence technology was utilized to reduce the instructor's workload as well as enhance learner engagement in online learning platforms. The idea of this paper is to use text-to-speech and speech-driven-face to automatically create a video lecture with the instructor's voice and face without recording the video. This approach not only gives the best display of lesson content of video lectures which contains slides (pdf format), instructor's voice, and face but also enables the modification of video lectures without recapture video. Furthermore, this approach allows learners to choose a video lecture with their desired instructor's voice and face. The experimental results validated our approach.