ABSTRACT

The current chapter presents an overview of how developments in automatic speech recognition (ASR) have been brought to bear on second language (L2) pronunciation training in the last two decades. The aim is to bridge the gap between speech technology and L2 pronunciation research with the ultimate goal of stimulating cooperation, which would benefit both research and practice. Since there are numerous publications on this topic, only a selection of illustrative studies is presented. The overview is mainly focused on studies that employ ASR prior to further analysis of pronunciation errors and is limited to research that addresses pronunciation training at segmental level. Although pronunciation assessment and pronunciation error detection are both mentioned and explained, the latter constitutes the central topic of this chapter. After a brief historical perspective and a discussion of critical issues, various ASR-based pronunciation training systems are discussed regarding the extent to which these are in line with theoretical insights from pronunciation research. The advantages of ASR-based pronunciation training systems for practice are considered, as well as the challenges that still need to be addressed. Finally, current limitations and possible future directions are discussed.