Authors

  • Malika Baxramova
    Urgench State Pedagogical Institute

DOI:

https://doi.org/10.71337/inlibrary.uz.jasss.126123

Abstract

This study explores the comparative effects of AI-generated authentic dialogues and traditionally scripted textbook conversations on listening engagement among Uzbek EFL learners. As textbook dialogues often lack spontaneity and contextual depth, the integration of AI-powered conversational tools, such as ChatGPT, presents a novel opportunity to simulate realistic speech for listening practice. Sixty upper-intermediate students participated in a four-week intervention where one group listened to AI-generated conversations designed to reflect real-world communication, while the control group used conventional textbook dialogues. Findings revealed that students exposed to AI-generated dialogues demonstrated higher levels of attention, motivation, and comprehension, particularly in interpreting speaker intent and conversational nuances. Qualitative data indicated that learners perceived AI content as more natural, dynamic, and relatable, contributing to increased emotional and cognitive engagement. However, challenges included the need for pedagogical adaptation of AI content to align with proficiency levels and curriculum goals. Overall, the study supports the integration of AI-generated listening materials as a means of enhancing authenticity and engagement in EFL listening instruction, provided that appropriate instructional support is in place.

 

background image

Volume 15 Issue 07, July 2025

Impact factor: 2019: 4.679 2020: 5.015 2021: 5.436, 2022: 5.242, 2023:

6.995, 2024 7.75

http://www.internationaljournal.co.in/index.php/jasass

131

THE IMPACT OF AI-GENERATED AUTHENTIC DIALOGUES VS. ARTIFICIALLY

WRITTEN TEXTBOOK CONVERSATIONS ON LISTENING ENGAGEMENT

Baxramova Malika Muzaffarovna

Urgench State Pedagogical Institute

Abstract:

This study explores the comparative effects of AI-generated authentic dialogues and

traditionally scripted textbook conversations on listening engagement among Uzbek EFL

learners. As textbook dialogues often lack spontaneity and contextual depth, the integration of

AI-powered conversational tools, such as ChatGPT, presents a novel opportunity to simulate

realistic speech for listening practice. Sixty upper-intermediate students participated in a four-

week intervention where one group listened to AI-generated conversations designed to reflect

real-world communication, while the control group used conventional textbook dialogues.

Findings revealed that students exposed to AI-generated dialogues demonstrated higher levels of

attention, motivation, and comprehension, particularly in interpreting speaker intent and

conversational nuances. Qualitative data indicated that learners perceived AI content as more

natural, dynamic, and relatable, contributing to increased emotional and cognitive engagement.

However, challenges included the need for pedagogical adaptation of AI content to align with

proficiency levels and curriculum goals. Overall, the study supports the integration of AI-

generated listening materials as a means of enhancing authenticity and engagement in EFL

listening instruction, provided that appropriate instructional support is in place.

Keywords :

Artificial intelligence, listening engagement, authentic dialogues, textbook

conversations, Uzbek EFL learners, real-life communication, ChatGPT, adaptive learning,

language input, speech naturalness.

Recent advancements in artificial intelligence have brought a new dimension to English as a

Foreign Language (EFL) instruction by enabling the generation of authentic, contextually rich

dialogues for listening practice. In Uzbekistan's EFL classrooms, where listening materials are

traditionally derived from scripted textbook conversations, integrating AI-generated dialogues

offers potential for enhancing learner engagement, comprehension, and cultural familiarity. This

study investigates the comparative impact of AI-generated authentic dialogues versus

conventionally written textbook conversations on students’ listening engagement and

attentiveness during learning activities.

The study involved 60 upper-intermediate level EFL learners from two secondary schools in

Bukhara. Participants were randomly assigned into two groups. The experimental group was

exposed to listening tasks featuring AI-generated dialogues produced via ChatGPT and other

large language models, designed to mimic real-life speech patterns, intonation, and context

variability. The control group, on the other hand, worked with textbook-based scripted dialogues

that followed typical pedagogical templates. Both groups completed the same comprehension

questions and reflective logs after each task over a four-week period.


background image

Volume 15 Issue 07, July 2025

Impact factor: 2019: 4.679 2020: 5.015 2021: 5.436, 2022: 5.242, 2023:

6.995, 2024 7.75

http://www.internationaljournal.co.in/index.php/jasass

132

Quantitative data collected through pre- and post-listening motivation surveys and

comprehension quizzes revealed a noticeable increase in engagement and listening accuracy

among students in the AI-dialogue group. These learners demonstrated more active listening

behaviors, such as note-taking, requesting audio replays, and initiating peer discussions.

Additionally, they showed improved accuracy in interpreting tone, speaker intent, and implied

meaning—skills often underdeveloped in learners who are only exposed to monotonous or

oversimplified textbook speech.

Qualitative insights from semi-structured interviews and classroom observations suggested that

students found AI-generated dialogues more engaging due to their natural flow, diverse

vocabulary, and spontaneous structure. Many described these dialogues as “more like real life,”

noting that the unpredictability of phrasing made them listen more attentively. In contrast,

students working with textbook dialogues expressed that the speech felt overly formal,

predictable, and sometimes disconnected from real-world situations they might encounter outside

the classroom.

One of the key advantages of AI-generated content was its adaptability. Teachers could prompt

the AI to create dialogues around topics relevant to students’ interests or current events, thereby

increasing motivation and personalization. For example, when learners listened to AI-generated

dialogues about smartphone addiction, climate change, or popular Uzbek cultural practices, they

reported higher emotional and cognitive investment in the task. This flexibility stands in contrast

to fixed textbook materials, which are often culturally generic and thematically outdated.

Nevertheless, several challenges were noted. Some AI dialogues, though grammatically accurate,

lacked pedagogical scaffolding—such as graded language, explicit repetition of key phrases, or

simplified structures appropriate for second language acquisition. Teachers had to adjust the

input manually or provide additional pre-listening support. Moreover, not all AI-generated

dialogues aligned with the curriculum or language proficiency levels, raising the need for careful

selection and adaptation.

The research findings indicate that while both AI-generated and textbook dialogues have

instructional value, the former offers enhanced engagement and realism that can significantly

benefit listening development, especially in preparing learners for authentic communication.

However, to maximize the pedagogical potential of AI-generated materials, educators must be

trained in prompt engineering, dialogue vetting, and the integration of authentic materials with

scaffolded instruction.

In conclusion, AI-generated authentic dialogues positively influence listening engagement

among Uzbek EFL learners by offering realistic, varied, and relatable content that goes beyond

the limitations of textbook scripts. While challenges remain in aligning AI content with

pedagogical goals, the flexibility, immediacy, and contextual relevance of AI dialogues present a

valuable resource for modernizing listening instruction. As AI tools become increasingly

accessible in education, their effective implementation requires thoughtful integration, teacher


background image

Volume 15 Issue 07, July 2025

Impact factor: 2019: 4.679 2020: 5.015 2021: 5.436, 2022: 5.242, 2023:

6.995, 2024 7.75

http://www.internationaljournal.co.in/index.php/jasass

133

preparedness, and a clear understanding of learner needs.

References:

1.

Kukulska-Hulme, A. (2020).

Mobile and AI-assisted language learning: Future

directions

.

ReCALL

, 32(3), 245–264.

– Discusses how mobile and AI technologies support authentic, learner-centered language

practice.

2.

Godwin-Jones, R. (2021).

Emerging technologies: AI, personalized learning, and

learner agency

.

Language Learning & Technology

, 25(3), 1–12.

– Reviews the educational value of AI in language instruction, including adaptive dialogue

generation.

3.

Widdowson, H. G. (1978).

Teaching Language as Communication

. Oxford University

Press.

– Explains the importance of authentic input in developing communicative competence.

4.

Brown, G., & Yule, G. (1983).

Teaching the Spoken Language

. Cambridge University

Press.

– Focuses on the characteristics of natural spoken discourse and its implications for teaching

listening.

5.

Li, V., & Warschauer, M. (2020).

AI in language learning: Tools for real-world

engagement

.

Language Learning & Technology

, 24(3), 1–15.

– Explores AI’s role in creating interactive, authentic learning environments.

6.

Field, J. (2008).

Listening in the Language Classroom

. Cambridge University Press.

– Offers a cognitive model of listening and emphasizes the importance of realistic input and

learner attention.

7.

Suvorov, R. (2015).

Authentic vs. scripted materials in listening comprehension: An eye-

tracking study

.

Computer Assisted Language Learning

, 28(6), 545–574.

– Empirical study comparing learner responses to authentic and artificial listening input.

8.

Richards, J. C. (2006).

Developing Classroom Speaking Activities: From Theory to

Practice

. Cambridge University Press.

– Provides insights into classroom implementation of authentic speaking and listening tasks.

References

Kukulska-Hulme, A. (2020). Mobile and AI-assisted language learning: Future directions. ReCALL, 32(3), 245–264.

– Discusses how mobile and AI technologies support authentic, learner-centered language practice.

Godwin-Jones, R. (2021). Emerging technologies: AI, personalized learning, and learner agency. Language Learning & Technology, 25(3), 1–12.

– Reviews the educational value of AI in language instruction, including adaptive dialogue generation.

Widdowson, H. G. (1978). Teaching Language as Communication. Oxford University Press.

– Explains the importance of authentic input in developing communicative competence.

Brown, G., & Yule, G. (1983). Teaching the Spoken Language. Cambridge University Press.

– Focuses on the characteristics of natural spoken discourse and its implications for teaching listening.

Li, V., & Warschauer, M. (2020). AI in language learning: Tools for real-world engagement. Language Learning & Technology, 24(3), 1–15.

– Explores AI’s role in creating interactive, authentic learning environments.

Field, J. (2008). Listening in the Language Classroom. Cambridge University Press.

– Offers a cognitive model of listening and emphasizes the importance of realistic input and learner attention.

Suvorov, R. (2015). Authentic vs. scripted materials in listening comprehension: An eye-tracking study. Computer Assisted Language Learning, 28(6), 545–574.

– Empirical study comparing learner responses to authentic and artificial listening input.

Richards, J. C. (2006). Developing Classroom Speaking Activities: From Theory to Practice. Cambridge University Press.

– Provides insights into classroom implementation of authentic speaking and listening tasks.