Design of Educational Software for Automatic Speech Recognition (ASR) Techniques

Abstract

Speech recognition has been an important subject for research, and it has come to a stage where it has been actively applied in a lot of industrial and consumer applications, overseas. However, speech recognition research is still in its infancy stage in Malaysia. The main reason is that speech recognition systems are highly complex and teaching students in this subject matter with the underlying technologies is a challenging task. Currently, some instructors use slide show presentations and white board in giving such courses. At the end of the course, students are not able to figure out the real output of the algorithms given. In this case, students are not exposed to the real technical systems and would easily get bored. This research is mainly on the improvement over the limitations and problems of the traditional teaching method in speech recognition by developing a set of interactive and practical education software to guide and assist students in studying, and performing experiments for speech recognition.