Developing a time- and cost-efficient method for second language (L2) oral proficiency measurement is one of the research topics that has attracted much attention in recent decades. The purpose of this study is to develop a computerized oral testing system for L2 Japanese using automatic speech recognition (ASR) technology. Two testing methods called elicited imitation (EI) and simulated speech (SS) are proposed to quantify L2 accuracy and fluency via ASR processing. This study also suggests systematic EI item creation leveraging corpus technology and discusses the effectiveness of the test items created through analyses of item difficulty. Further, refinement of the EI grading system is described through a series of statistical investigations. For SS, this study reports the five most influential L2 fluency features identified through machine learning and proposes a method to yield individual SS scores with these features based on previous studies. Lastly, several methods to combine the EI and SS scores are presented to estimate L2 oral proficiency of Japanese.

The objective and automatic grading of oral language tests has been the subject of significant research in recent years. Several obstacles lie in the way of achieving this goal. Recent work has suggested a testing technique called elicited imitation (EI) can be used to accurately approximate global oral proficiency. This testing methodology, however, does not incorporate some fundamental aspects of language such as fluency. Other work has suggested another testing technique, simulated speech (SS), as a supplement to EI that can provide automated fluency metrics. In this work, I investigate a combination of fluency features extracted for SS testing and EI test scores to more accurately predict oral language proficiency. I also investigate the role of EI as an oral language test, and the optimal method of extracting fluency features from SS sound files. Results demonstrate the ability of EI and SS to more effectively predict hand-scored SS test item scores. I finally discuss implications of this work for future automated oral testing scenarios.