Article Per Year (5 Year)
p-Index From 2021 - 2026
0
P-Index
This Author published in this journals
All Journal Educational Journal of Learning Technology
Fera Favirotus Siyam
Universitas Darussalam Gontor
Computer Science & IT Education Languange, Linguistic, Communication & Media Social Sciences Other

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Accuracy Analysis of Text-to-Speech (TTS) on Chatbot GPT for Arabic Conversation and Grammatical Mapping Fasya Asy Shyifaauzzahra; Moh. Ismail; Fairuz Subakir Ahmad; Rosendah Dwi Maulaya; Ulfa Muna Kamila; Fera Favirotus Siyam; Annisa Avilya
Educational Journal of Learning Technology Vol. 4 No. 2 (2027): Educational Journal of Learning Technology (Oktober-Januari 2027)
Publisher : South Sulawesi Education Development (SSED)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58230/edutech.v4i2.144

Abstract

This qualitative study evaluates ChatGPT-4o's spoken Arabic capabilities and grammatical mapping accuracy across two distinct devices using a single foundational Arabic sentence prompt ("اللغة العربية هي لغة الجنة"). The evaluation criteria encompassed diacritical precision (harakat), phonetic articulation (makhraj), syntactic comprehension (nahw), and I'rab parsing. We cross-verified data generated through structured dialogues against Al-Nahw Al-Wadih and validated it with an Arabic linguistics expert. The empirical findings indicate that ChatGPT-4o achieved uniform excellence, scoring 4 out of 4 in diacritical rendering and phonetic articulation across both devices. However, in systematic syntactic parsing (I'rab), the model achieved measured grammatical accuracy rates of 78.4% on Device I using an Arabic interface and 74.5% on Device II using an English interface. These findings show that while AI can support Arabic conversational practice and basic linguistic inquiries, its structural parsing performance remains constrained by default system language configurations, hardware variations, and overall sentence complexity.
Co-Authors Annisa Avilya Fairuz Subakir Ahmad Fasya Asy Shyifaauzzahra Moh. Ismail Rosendah Dwi Maulaya Ulfa Muna Kamila