基于CSE口译量表的口译自动评分路径探索  

Automatic scoring based on the Interpreting Scale of China’s Standards of English Language Ability

作　　者：王巍巍[1] 王轲 张昱琪 WANG Weiwei;WANG Ke;ZHANG Yuqi

机构地区：[1]广东外语外贸大学高级翻译学院,广东广州510420 [2]阿里巴巴达摩院机器智能实验室,北京100005

出　　处：《外语界》2022年第2期80-87,共8页Foreign Language World

基　　金：2019年国家社科基金青年项目“中国英语能力等级量表口译自评量表的验证与应用研究”(编号19CYY053)的阶段性成果。

摘　　要：本研究基于《中国英语能力等级量表》(CSE)的口译量表,运用人工智能(AI)技术研发了可以应用于口译教、学、测的口译自动评分系统。初步数据显示,AI算法模型1.0和人类评审在同质口译学生群体中打分的Pearson相关系数为0.95。对于不同应用场景中的异质口译学生群体,在人类评审一致性较高的情况下,AI模型的分值在一定范围内波动。研究认为,利用AI算法模型实现口译量表工具化、开展口译自动评分具有一定的可行性。Based on the Interpreting Scale of China’s Standards of English Language Ability(CSE)and artificial intelligence(AI)technology,the current study develops an automatic scoring system for interpreting,hopefully applicable to interpreting teaching,learning,and testing.Preliminary data shows that the Pearson coefficient of the AI algorithm model 1.0 and human reviewers is 0.95 in the homogeneous interpreting student population.In the case of heterogeneous interpreting students in different application scenarios and high consistency among human reviewers,the scores of AI model fluctuate in a certain range.The study concludes that it is feasible to use AI algorithm models to apply CSE-Interpreting Scales in automatic scoring of interpreting.

关 键 词：CSE口译量表 口译自动评分 算法 

分 类 号：H319[语言文字—英语]