面向口译教学的视频语料资源库深度开发机制研究

Deep processing mechanisms of the conversion of video corpora to speech repository for interpreting teaching

语料库在语言研究、外语教学、翻译学研究、词典编纂等领域已有广泛应用。与此同时,语料库,尤其是视频语料库,在口译教学中的应用价值日益受到关注。基于语料库开发面向口译教学的语料资源库必须秉持由研究驱动向教学驱动转化的理念,且需要综合考虑素材、职业、教学法、语境等因素。在实践层面,本文以欧盟委员会"虚拟现实口译"项目的视频语料资源库为例,从素材选取、语料切分、信息标注与场景模拟四个方面阐述其深度开发机制,并探讨其对我国口译教学语料资源建设的启示。

Corpora have been widely used in language research, foreign language teaching, translation studies and dictionary compilation. Meanwhile, corpora, and video corpora in particular, have been attracting more and more attention because of their potential application in interpreting teaching. There must be a shift from research-driven to teaching-driven as for the conversion of video corpora to speech repository for interpreting teaching. Moreover, material, professional, pedagogical and contextual factors all need to be taken into consideration in the conversion. In practice, the video corpora in the project of "Interpreting in Virtual Reality"(IVY) funded by the European Commission offer a good example, as demonstrated in speech selection, division, annotation and scenario simulation. The implications of IVY are also discussed for the development of speech resources for interpreting teaching in China.