AI-Assisted Contrastive Phonological Study of Arabic and Mandarin Consonants.

Read full text
Authors: Asmaa Houssein Abdelmajeed, Sameh Alansary
Cited by: 2
References: 2

Abstract

This study aims to make a comparative analysis between the phonological systems of consonants in Modern Standard Arabic and Mandarin Chinese. Through this contrastive analysis of the phonological and phonetic structures concentrating on the articulation, distinctive features, and phonetic symbols of sounds, it explores the similarities and differences between the two systems. In addition, that it discusses how Artificial Intelligence (AI) prompts can help in phonological errors pattern. The data was selected from 13 Chinese students who learn Arabic level and their levels was upper-intermediate and advanced, according to the CEFR scale (Common European Framework of Reference for Languages), the data was recorded from their spontaneous speech and their permissions were taken after recording to make sure they are talking naturally and also for some ethical reasons. The results show notable differences between the Arabic and Chinese sound systems, particularly in the manner, place of articulation of specific consonants and voicing. These significant differences assist in explaining the challenges faced by Chinese learners while learning Arabic as a second foreign language. In Arabic language there are sounds absent in Chinese, such as the dental and emphatic consonants, while Mandarin has also sounds not found in Arabic phonology, such as retroflex and alveo-palatal consonants. Moreover, there are some pronunciation errors related to the concept of voicing, which differs between the two languages. The results showed that a strong positive correlation, which was observed between AI and human scores (r = 0.81, 95 % CI [0.72, 0.88], p < 0.001).

Fields of Study & Subjects

Phonetics and Phonology Research Speech Recognition and Synthesis Emotion and Mood Recognition

Publication details

Journal / source
Procedia Computer Science
Publisher
Elsevier BV
Volume
275
Pages
56–65
Publication date
1 January 2026
Language
EN