Integrating Speech Technology into the iSTART-Early Intelligent Tutoring System

Speech technology (automated speech recognition – ASR and text-to-speech) offers great promise in the field of automated literacy and reading tutors for children. Students in third and fourth grades struggle with generating longer strings of text on a QWERTY keyboard because they still “hunt and peck” for the letters and symbols rather than typing fluently. Thus, in addition to reading comprehension, students’ performance is limited by their ability to translate their ideas into language and then transcribe those words into written text. Fourth grade students produce fewer words and recall less when typing or writing a response relative to speaking. Hence, speech technology is a crucial component in the development of iSTART-Early, an intelligent tutoring system that aims to provide online, automated reading strategy instruction and practice to improve deep comprehension for third and fourth graders. This paper discusses the key components and features of the speech technology incorporated into iSTART-Early. We also discuss some initial findings from pilot studies conducted with adults and youth for this ASR integrated tutoring system. Finally, we discuss considerations for future development of speech technology and integration with intelligent tutoring systems.