Page 606 - ISC PROCEEDINGS 21.4

The scores regarding pronunciation also have positively changed (M = 60.12 SD =
7.45 during the pre-assessment and M = 78.30 SD = 6.98 during the post-assessment).
The paired-sample t-test was used to assert that the statistical significance of the
improvement was significant, t(49) = 10.26, p < 0.001, with a large effect size (d = 1.45).
This observation indicates that speech recognition aids that were enabled by AI
contributed significantly to the increase in pronunciation accuracy among learners.
Table 2. Pronunciation assessment results

Assessment N Mean Score Standard Deviation Mean Difference

Pre-Assessment 50 60.12 7.45 —
Post-Assessment 50 78.30 6.98 +18.18
Source: Author
The results of the pronunciation assessment show that the accuracy of the
pronunciation of the learners has greatly improved. The post-assessment scores went up
to 78.30 as compared to 60.12, which indicated that AI-type speech recognition resources
could offer effective feedback resulting in an improvement in the pronunciation abilities
of students.
6.3. Student interests and viewpoints
The qualitative information obtained in case of student questionnaires and
classroom observations offered additional information on the experience of learners
using AI-based language learning tools. On balance, most of the participants were positive
in the process of studying the language with the help of AI technologies. According to
very many learners, language learning became more interactive and engaging with the
help of AI applications. Digital tools provided them with an opportunity to train their
vocabulary and pronunciation in many different activities, including quizzes, speech
recording exercises, and interactive activities. These characteristics served to sustain the
interest of the learners and made them take more time to practice English. Moreover, the
respondents enjoyed instant feedback of the AI systems. Unlike the conventional
classroom environments where feedback could be delayed, AI technologies gave the
learners an opportunity to know their mistakes and rectify them immediately. This form
of instant feedback made learning more effective and allowed the learners to track their
progress. Nonetheless, other participants also reported certain small difficulties in the use
of AI tools, including technical problems or the lack of accuracy in speech recognition.
Regardless of these shortcomings, the majority of learners accepted the idea that AI
technologies were a worthy addition to language training.

Table 3. Students’ perceptions of AI-based language learning tools
Agree Neutral Disagree
Statement
(%) (%) (%)
AI tools help improve vocabulary learning 82 10 8
AI pronunciation feedback is useful 85 9 6
AI tools increase learning motivation 79 12 9
AI tools make language learning more 84 8 8
interactive

Source: Author

605