Abstract:

We construct a spoken interaction (SI) corpus of Mandarin Chinese in this paper, which is the first hierarchical corpus of SI in China. This corpus is a valuable language resource in which the spoken interactions among more than 1200 students are recorded, and the whole duration time is more than 3000 minutes. The students range from Grade 1 in primary school to Grade 3 in high school. This corpus provides researchers with materials with transcriptions and annotations, by which the quantitative analysis for SI can be realized. In addition, the materials are graded according to Conversation Analysis（CA）rather than task levels, providing reference for researchers. So the textbook compiling and the establishment of SI grading standards can both benefit from this corpus.

Key words: spoken interaction;graded corpus;grading standard