Transcending the display size: the case for speech interaction in educational applications

Speech is the most natural form of communication that humans employ, and is one of the main modalities through which we acquire and share knowledge. Moreover, speech is used not only to deliver knowledge, but as a modality that supports learning, such as student-teacher interactions around printed materials. During the past decade, we have witnessed significant advances mainly in preserving spoken educational materials, from informal how-to videos to full academic lectures being stored and available through a variety of online channels. Unfortunately, there is proportionately less research on enabling access to such multimedia knowledge repositories (e.g. searching, indexing) or on facilitating spoken, natural interaction between learners and digital interactive media (such as automated tutors or interactive learning resources). By enabling speech as a modality, learners become less constrained by the physical properties of the educational materials and can interact more naturally with the educational software, be it in the form of a mobile language assistant, a desktop-based online lecture browsing system, or a mixed-reality serious gaming system. In this paper I present examples of such recent research on improving the way we interact with educational resources through speech and natural language, and make the case for the need to conduct further research in this area.Peer reviewed: YesNRC publication: Ye