Teacher Education in CALL   

< Table of Contents

Chapter 8. 
Teaching the Creation of Software That Uses Speech Recognition

Maxine Eskenazi and Jonathan Brown, Carnegie Mellon University, USA

This chapter describes the content and rationale for the first part of a Carnegie Mellon University course on the creation of software that uses language technologies and in particular, speech recognition. This course brings the students from the theory they have seen in other courses to the actual design of small portions of systems. The high level goal of the course is to teach what constitutes good language learning software, and how students can create their own. Students learn about human-computer interface issues, intelligent tutoring methods and about how to use language technologies in a real piece of tutoring software. The chapter discusses the overall structure of the course, and then focuses on the part in which students learn the basics of language technologies for CALL systems and produce a language learning application that uses speech recognition.

References available online

CMU Pronouncing Dictionary. Available at  http://www.speech.cs.cmu.edu/cgi-bin/cmudict/.

Sphinx Knowledge Base Tools. Available at http://www.speech.cs.cmu.edu/tools/.

Sphinx Project Page. Available at http://www.speech.cs.cmu.edu/sphinx/.

Sphinx2-CTAT Connection Utilities. Available at  http://www.cs.cmu.edu/~jonbrown/Sphinx2-CTAT/.


< Table of Contents

Updated September 12, 2006