基于语音识别技术的口译实证研究——成果、问题与展望  

Empirical Studies on Interpreting with Speech Recognition Technology:Progress,Challenges and Future Prospects

作　　者：钟林苹 马星城 ZHONG Linping;MA Xingcheng(Southeast University)

机构地区：[1]东南大学

出　　处：《翻译界》2025年第1期116-131,共16页Translation Horizons

基　　金：教育部人文社会科学研究青年基金项目“基于眼动数据库的英汉双向同声传译认知机制多维度研究”(23YJC740047);江苏省社会科学基金项目“语料库辅助的汉英口译认知机制研究”(21CYY06)的阶段性研究成果。

摘　　要：随着人工智能技术的迅猛发展,计算机辅助翻译已成为翻译研究的前沿热点。语音识别是计算机辅助口译的核心技术,但其在实际应用中的研究仍有待深入。本文旨在探讨语音识别技术在口译领域的应用现状,系统梳理了2013—2023年国内外相关实证研究,归纳了主要研究主题、方法和结论。基于口译活动的认知特殊性与情境复杂性,本文从认知科学与认知翻译学的实证—实验研究设计标准出发,剖析了现有研究中存在的问题与局限。通过本研究,我们期望加深学界与口译行业对语音识别技术的理解与思考,促进技术赋能下口译实践、教学与研究的良性互动,为未来研究方向提供启示。With the rapid advancement of artificial intelligence,computer-assisted translation has emerged as a cutting-edge focus in translation studies.Speech recognition serves as the core technology for computer-assisted interpreting,yet its practical application remains understudied.This paper aims to explore the current state of speech recognition technology in interpreting by systematically reviewing relevant empirical studies conducted both domestically and internationally from 2013 to 2023.We synthesize the main research themes,methodologies,and findings.Considering the unique cognitive processes and contextual complexities in interpreting,we analyze the problems and limitations in existing research based on the standards of empirical-experimental research design in cognitive science and cognitive translation studies.Through this study,we hope to deepen the understanding and reflection on speech recognition technology in academia and the interpreting industry,promote the positive interaction among interpreting practice,teaching,and research empowered by technology,and provide insights for future research directions.

关 键 词：计算机辅助口译 语音识别 口译技术 实证研究 认知加工 

分 类 号：H059[语言文字—语言学]