生成式人工智能赋能英语口语教学的实证研究
Empowering English Speaking Instruction with Generative AI: An Empirical Study
摘要: 研究以60名英语专业学生为对象,考察生成式人工智能在口语训练中的用法及其对口语产出的影响。实验组(30名)使用豆包AI辅助训练,对照组(30名)采用自主录音回听训练,借助雅思口语Part 2、Part 3任务开展8周共4次训练,从词汇复杂度、句法复杂度、准确度与流利度四个维度分析口语产出,并结合访谈、反思日志与有声思维报告作主题分析。结果显示:实验组词汇复杂度随训练显著上升(F = 13.63, p < 0.001),准确度显著下降(F = 91.80,p < 0.001,偏η2 = 0.760),实验组句法复杂度先升后降,两组流利度均无显著改善,实验组在复杂度与准确度之间的取舍更明显。质性分析归纳出学习陪伴者、反馈纠错者、学习示范者、输出与反馈重述闭环、学习监督者五类人机交互模式,前四类已经出现,监督者角色尚未形成,说明AI主要提供练习场合、错误反馈和语言范例,训练仍由学习者推进。研究认为,AI能否改善口语教学,主要取决于训练环节如何组织,而不取决于工具本身。
Abstract: This study examines how generative artificial intelligence (GAI) is used in speaking practice and how it affects the quality of learners’ oral production. Sixty English majors were assigned to an experimental group (n = 30, Doubao AI-assisted training) and a control group (n = 30, self-recording and playback training). Over eight weeks and four training sessions based on IELTS Speaking Parts 2 and 3, oral productions were analysed in terms of lexical complexity, syntactic complexity, accuracy and fluency, and interviews, reflective journals and think-aloud reports were analysed thematically. Lexical complexity rose significantly across sessions in the experimental group (F = 13.63, p < 0.001), while accuracy fell significantly (F = 91.80, p < 0.001, partial η2 = 0.760). Syntactic complexity rose and then declined in the experimental group, and fluency did not improve significantly in either group; the experimental group showed a clearer trade-off between complexity and accuracy. Five patterns of human-AI interaction were identified: companion, corrector, demonstrator, an output-feedback-restatement loop, and supervisor. The first four were observed, whereas the supervisory role did not emerge, indicating that AI mainly provides practice settings, error feedback, and model language, while the training is still driven by learners. The study concludes that whether GAI improves speaking instruction depends mainly on how training is organised rather than on the tool itself.
文章引用:叶倾宸, 刘思睿, 姚佳欣, 尹浩, 古明. 生成式人工智能赋能英语口语教学的实证研究[J]. 教育进展, 2026, 16(11): 1092-1100. https://doi.org/10.12677/ae.2026.16112478

参考文献

[1] 郭雄如. 英语口语教学的语言环境和教学法[J]. 韶关大学学报(社会科学版), 1996, 17(1): 108-111.
[2] 张琳, 金艳. 基于交互观的口语互动能力构念界定[J]. 外语学刊, 2016(2): 103-108.
[3] 吴坚豪, 周婉婷, 曹超. 生成式人工智能技术赋能口语教学的实证研究[J]. 中国电化教育, 2024(4): 105-111.
[4] Canale, M. (1980) Theoretical Bases of Communicative Approaches to Second Language Teaching and Testing. Applied Linguistics, 1, 1-47.
https://doi.org/10.1093/applin/1.1.1
[5] 卢宇, 余京蕾, 陈鹏鹤, 等. 生成式人工智能的教育应用与展望——以ChatGPT系统为例[J]. 中国远程教育, 2023, 43(4): 24-31, 51.
[6] 潘海英, 李佳佳, 马新月. 生成式人工智能在外语教学中的应用与研究: 一项系统性综述[J]. 外语教学与研究, 2026, 58(2): 250-262.
[7] 郑春萍, 于淼, 郭智妍. 人工智能在语言教学中的应用研究: 回顾与展望[J]. 外语教学, 2024, 45(1): 59-68.
[8] 王立群. 我国大规模英语测试的伦理体系建构[J]. 中国考试, 2019(11): 7-13.
[9] Qiao, H. and Zhao, A. (2023) Artificial Intelligence-Based Language Learning: Illuminating the Impact on Speaking Skills and Self-Regulation in Chinese EFL Context. Frontiers in Psychology, 14, Article 1255594.
https://doi.org/10.3389/fpsyg.2023.1255594
[10] 孙璐. 生成式人工智能赋能英语专业研究性教学: 人机协同模式探索[J]. 外语电化教学, 2026(1): 10-15, 106.
[11] Ericsson, K.A. and Simon, H.A. (1993) Protocol Analysis: Verbal Reports as Data. 2nd Edition, The MIT Press.
https://doi.org/10.7551/mitpress/5657.001.0001
[12] 金霞. 工作记忆容量限制对二语学习者口语产出的影响[J]. 外语教学与研究, 2012, 44(4): 523-535, 640.
[13] 韩亚文. 工作记忆容量对中国英语学习者口语准确度、复杂度和流利度的影响[J]. 外语教学, 2015, 36(5): 65-68.
[14] Skehan, P. (2009) Modelling Second Language Performance: Integrating Complexity, Accuracy, Fluency, and Lexis. Applied Linguistics, 30, 510-532.
https://doi.org/10.1093/applin/amp047
[15] Mehnert, U. (1998) The Effects of Different Lengths of Time for Planning on Second Language Performance. Studies in Second Language Acquisition, 20, 83-108.
https://doi.org/10.1017/s0272263198001041
[16] Swain, M. (1995) Three Functions of Output in Second Language Learning. In: Cook, G. and Seidlhofer, A., Eds., Principle and Practice in Applied Linguistics, Oxford University Press, 125-144.
[17] Biber, D., Gray, B. and Poonpon, K. (2011) Should We Use Characteristics of Conversation to Measure Grammatical Complexity in L2 Writing Development? TESOL Quarterly, 45, 5-35.
https://doi.org/10.5054/tq.2011.244483
[18] Long, M.H. (1996) The Role of the Linguistic Environment in Second Language Acquisition. In: Ritchie, W.C. and Bhatia, T.K., Eds., Handbook of Second Language Acquisition, Elsevier, 413-468.
https://doi.org/10.1016/b978-012589042-7/50015-3
[19] Bronfenbrenner, U. (2005) Making Human Beings Human: Bioecological Perspectives on Human Development. Sage.
[20] Kohnke, L. (2023) A Pedagogical Chatbot: A Supplemental Language Learning Tool. RELC Journal, 54, 828-838.
https://doi.org/10.1177/00336882211067054
[21] Huang, W., Hew, K.F. and Fryer, L.K. (2022) Chatbots for Language Learning—Are They Really Useful? A Systematic Review of Chatbot‐supported Language Learning. Journal of Computer Assisted Learning, 38, 237-257.
https://doi.org/10.1111/jcal.12610
[22] 文秋芳. 构建“产出导向法”理论体系[J]. 外语教学与研究, 2015, 47(4): 547-558.
[23] 张文娟. AI赋能产出导向法教材使用的课堂实践研究[J]. 现代外语, 2025, 48(5): 679-689.
[24] 焦建利, 陈婷. 大型语言模型赋能英语教学: 四个场景[J]. 外语电化教学, 2023(2): 12-17, 106.
[25] Lin, C.J. and Mubarok, H. (2021) Learning Analytics for Investigating the Mind Map-Guided AI Chatbot Approach in an EFL Flipped Speaking Classroom. Educational Technology & Society, 24, 16-35.