Visual Speech in Technology-Enhanced Learning

摘要: This thesis investigates the use of synthetic talking heads, with lip, tongue and face movements synchronized synthesized or natural speech, in technology-enhanced learning. work applies heads a speech tutoring application for teaching English as second language. Previous studies have shown that perception is aided by visual information, but more research needed to determine effectiveness visualization articulators pronunciation training. explores whether not technology can give an improvement learning pronunciation. This techniques audiovisual synthesis, using both viseme-based data-driven approaches implement multiple heads. Intelligibility found be intelligible than audio alone, head was viseme-driven implementation. The are applied pronunciation-training application, which evaluated second-language learners investigate benefit User trials explored efficacy software demonstrating /b/–/p/ contrast English. The results indicate showed listening after software, while compared auditory training alone varied between individuals. evaluations were perceived helpful pronunciation, positive feedback on system suggests could useful addition traditional methods.