Literacy is a prerequisite to participate in our knowledge society and reading skills are essential for school success. Developing reading skills requires intensive practice. Guided reading aloud is an important practice form which currently is only possible to a limited extent, because it requires individual attention by teachers. As increasing numbers of foreign students in Dutch schools aggravate this problem, schools are looking for ways of providing more opportunities to practice guided reading aloud. Educational software that incorporates Automatic Speech Recognition (ASR) is a possible solution because it can provide instantaneous, automated feedback on reading aloud.
In this project, we investigate how reading software with ASR can best be developed and implemented in the school practice and at home. A design research methodology is employed in which ASR-based reading software is designed based on teachers’, students’ and parents’ needs and requirements. Design ideas and prototypes are repeatedly tested and evaluated with students, teachers and parents throughout the design process. A special focus is on suitable forms of feedback on reading aloud.
To investigate the effects of using ASR-based reading software on students’ reading skills, two randomized pretest-posttest control-group experiments are carried out, one in the school practice and one at home. Differential effects of the software and feedback types for different types of students are investigated, offering valuable information for personalized learning environments.
This research will provide important practical insights on the possibilities and effectiveness of using ASR-based software in reading education, and will advance scientific knowledge on reading development.