自动语音识别技术对大学英语学生发音准确度的影响  

The Impact of Automatic Speech Recognition Technology on the Pronunciation Accuracy of College English Students

作　　者：范园 Fan Yuan(Lingnan Normal University,Zhanjiang 524037,China)

机构地区：[1]岭南师范学院,广东湛江524037

出　　处：《黑河学院学报》2025年第3期83-85,179,共4页Journal of Heihe University

基　　金：2022年广东省本科高校教学质量与教学改革工程建设项目“基于ASR技术提高大学英语学生口语准确度研究与改革实践”(粤教高函〔2023〕4号)。

摘　　要：自动语音识别是一种有效处理口语数据并将其转化为书面语的技术,通过对听得见但看不见的声音进行可视化处理,使其呈现更加直观的形式,并为学生提供有关其发音的反馈。对17名大二年级学生进行了为期4周的实验,通过对4次口语朗读测试的前测、后测成绩的对比分析,发现学生的英语发音准确度有了显著的提高。参与者在采访中表示基于自动语音识别技术的语音练习为其创设了安全的练习空间,有效地降低了表达焦虑。Automatic Speech Recognition(ASR)is a technology that effectively processes spoken data and converts it into written texts.By visualizing audible but invisible sounds,it’s found that ASR can offer a more intuitive representation and provide automated feedback on students’pronunciation.This study conducts a 4-week experiment with 17 sophomore students,analyzing the differences between pre-test and post-test scores from four oral reading assessments.The results indicate a significant improvement in students’English pronunciation accuracy after the ASR-based intervention.Participants report that the ASR-based practice can create a safe space for them,effectively reducing their anxiety about speaking.

关 键 词：自动语音识别 发音准确度 表达焦虑 安全空间 

分 类 号：G642[文化科学—高等教育学]