数智驱动背景下的语音识别技术在英语口语教学中的应用研究

A Study on the Application of Speech Recognition Technology in English Speaking Instruction under the Data-Intelligence-Driven Context

摘要: 全球化与智慧教育背景下，大学英语口语教学受大班制反馈不足、学生自主练习低效等问题制约。本文定位为基于系统性文献回顾的教学模式构建与应用提案，基于纠正性反馈理论，系统梳理国内外相关研究与典型案例，分析数智驱动下语音识别技术在英语口语“识别–教学–反馈”环节的应用。研究表明，语音识别技术可实现段音规模化检测、客观评估与即时反馈，缓解教师反馈压力，支撑“课前–课中–课后”学习；但在超段音检测、错误原因解释及情感支持上存在局限，需与教师协同形成混合反馈模式。本文通过综述明确语音识别技术的应用价值与局限，为大学英语口语教学智能化转型提供参考，也为纠正性反馈理论在新技术场景的应用拓展思路。

Abstract: In the context of globalization and smart education, college English speaking instruction faces challenges such as limited feedback in large-class settings and low efficiency in students’ autonomous practice. This study positions itself as a teaching model construction and application proposal based on a systematic literature review. Based on corrective feedback theory, this paper systematically reviews relevant domestic and international research and representative cases, analyzing the application of Automatic Speech Recognition (ASR) technology in the “recognition-instruction-feedback” stages of English speaking under intelligent data-driven conditions. The study finds that ASR can achieve large-scale phoneme-level detection, objective evaluation, and instant feedback, thereby alleviating teachers’ feedback burden and supporting learning before, during, and after class. However, limitations remain in suprasegmental feature detection, error cause interpretation, and emotional support. Thus, a hybrid feedback model integrating teacher participation is necessary. Through this review, the paper clarifies the value and limitations of ASR in college English speaking instruction, providing references for the intelligent transformation of English teaching and expanding the application scope of corrective feedback theory in emerging technological contexts.