An Application Study on AI Educational Robots in Spoken English Exercises of Chinese Primary Schools

Yang Feng, Yu Yanan, Lina Cao


China’s primary schools offer limited English courses, and the society lacks the environment for naturally acquiring English in everyday life. Students typically have weak spoken English abilities and inadequate application of English. With the aim of addressing this issue, 79 fifth-graders from China’s Hangzhou L elementary school participated in a one-semester AI-assisted English-speaking practice experiment. The control class practiced spoken English by reading English texts aloud, whereas the experimental class practiced for 30 minutes a day using the “AI educational robots + graded picture books + role play” approach. According to the results of the experiment’s post-test, Chinese primary school students regarded the experimental class’s acquisition mode to be highly appealing, this approach was well accepted by both students and parents and brought enthusiasm and good effect of spoken English exercise. The experimental class’s average daily reading time for English role-play reading grew by about 30 minutes, the amount of reading increased by five times, the amount of time spent watching cartoons and playing video games fell by nearly 28 minutes, and the spoken English score climbed by 37 points, representing an increase of 82% when compared to the control class; Additionally, the standard level of pronunciation and intonation has increased by two grades, from “poor” to “good,” and the English final exam scores have increased by roughly 8%. However, there has not been a considerable change in the aforementioned control class indicators. This AI-assisted second language practice technique is affordable, efficient, and helpful and has good implications for second language acquisition in other countries.

Full Text:




  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © SCHOLINK INC.  ISSN 2372-9740 (Print)  ISSN 2329-311X (Online)