An online system for entering and annotating non-native Mandarin Chinese speech for language teaching

Abstract

This thesis describes the design and implementation of an intuitive online system for the annotation of non-native Mandarin Chinese speech by native Chinese speakers. This system will allow speech recognition researchers to easily generate a corpus of labeled non-native speech. We have five native Chinese speakers test the annotation system on a sample bank of 250 Chinese utterances and observe fair to moderate inter-rater agreement scores. In addition to giving us a benchmark for inter-rater agreement, this also demonstrates the feasibility of having remote graders annotate sets of utterances. Finally, we extend our work to Chinese language instruction by creating a web-based interface for Chinese reading assignments. Our design is a simple, integrated solution for completing and correcting of spoken reading assignments, that also streamlines the compilation of a corpus of labeled non-native speech for use in future research.