Melola Discovery

← Melola Discovery

Akses terbuka Jurnal SINTA 1

Assessing AI-Powered Translation Quality: Insights from the Translation of the Farewell Sermon

IJoLE: International Journal of Language Education · Vol. 8 (2025)

DOI
10.26858/ijole.v8i4.70034
ISSN
2548-8465, 2548-8457
Terbit
2025-01-01
Penerbit
Universitas Negeri Makassar
Bahasa
eng
Lisensi
creativecommons.org/

Abstrak

Artificial Intelligence (AI) powered translation influxes in the translation industry. Yet, questions have arisen about the quality of AI-powered translation systems (AIPTSs). Therefore, this study aims to evaluate the translation quality generated by ChatGPT for the Farewell Sermon. Using House's (2015) model of translation quality assessment (TQA), the study analyzed the Farewell Sermon focusing on text, register analysis of field, tenor mode, and genre. The language/text analysis revealed that AIPTSs produced lexical and syntactic inaccuracies that hindered capturing the sermon's intended religious voice. In register, Field, the analysis pointed out that the AIPTSs captured the broad themes of the Farewell Sermon, but often missed subtleties in religious terminology. AIPTSs also adopted superficial literal translation of the sentence structures. In Tenor, the analysis revealed that the Prophet's role as a moral and spiritual guide was somewhat diluted in the translation. Likewise, AIPTSs struggled to maintain the Prophet's authoritative and compassionate stance. The AIPTSs maintained general coherence but faced challenges with cohesive devices such as conjunctions, pronouns, and references. Furthermore, the microanalysis revealed an amount of 81 overt translation errors that the AIPTSs committed while translating the Farewell Sermon into English, the most frequent were "creative translation" with 24 errors, "not translated" with 15 errors, "distortion of meaning" with 9 errors, "slight change in meaning" with 8 errors and "breach of the SL system" with 7 errors. "Significant change in meaning" and "cultural filtering" were less noticeable with 6 and 4 errors, respectively. Additionally, the analysis revealed that House's (2015) model is to some extent suitable for assessing the quality of AIPTSs, one type of error was technical and beyond the model focus though. The researchers called it "software intervention". Software interventions can either positively or negatively affect the translation depending on whether the error improves readability without distorting meaning or inaccuracies. The study recommends the importance of tuning the AIPTSs applications to prevent the inducing of technical systems

Jaringan kolaborasi para penulis

6 penulis · 9 tautan kepenulisan bersama · 8 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • King Khalid University
  • Hadhramout University
  • Universitas Pattimura
  • University of Rochester
  • Hasanuddin University
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Muhammad Ridha Anugrah LatiefHasanuddin University1
Karolis AnaktototyUniversitas Pattimura1
Mansye SekewaelUniversity of Rochester1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) http://ojs.unm.ac.id/ijole/oai, pertama di-harvest 8 Nov 2026, terakhir terlihat 16 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Assessing AI-Powered Translation Quality: Insights from the Translation of the Farewell Sermon</dc:title>
	<dc:creator>Metwally, Amal Abdelsattar</dc:creator>
	<dc:creator>Bin-Hady, Wagdi Rashad Ali</dc:creator>
	<dc:creator>Asiri, Eisa</dc:creator>
	<dc:subject xml:lang="en-US"></dc:subject>
	<dc:subject xml:lang="en-US"></dc:subject>
	<dc:subject xml:lang="en-US"></dc:subject>
	<dc:description xml:lang="en-US">Artificial Intelligence (AI) powered translation influxes in the translation industry. Yet, questions have arisen about the quality of AI-powered translation systems (AIPTSs). Therefore, this study aims to evaluate the translation quality generated by ChatGPT for the Farewell Sermon. Using House's (2015) model of translation quality assessment (TQA), the study analyzed the Farewell Sermon focusing on text, register analysis of field, tenor mode, and genre. The language/text analysis revealed that AIPTSs produced lexical and syntactic inaccuracies that hindered capturing the sermon's intended religious voice. In register, Field, the analysis pointed out that the AIPTSs captured the broad themes of the Farewell Sermon, but often missed subtleties in religious terminology. AIPTSs also adopted superficial literal translation of the sentence structures. In Tenor, the analysis revealed that the Prophet's role as a moral and spiritual guide was somewhat diluted in the translation. Likewise, AIPTSs struggled to maintain the Prophet's authoritative and compassionate stance. The AIPTSs maintained general coherence but faced challenges with cohesive devices such as conjunctions, pronouns, and references. Furthermore, the microanalysis revealed an amount of 81 overt translation errors that the AIPTSs committed while translating the Farewell Sermon into English, the most frequent were "creative translation" with 24 errors, "not translated" with 15 errors, "distortion of meaning" with 9 errors, "slight change in meaning" with 8 errors and "breach of the SL system" with 7 errors. "Significant change in meaning" and "cultural filtering" were less noticeable with 6 and 4 errors, respectively. Additionally, the analysis revealed that House's (2015) model is to some extent suitable for assessing the quality of AIPTSs, one type of error was technical and beyond the model focus though. The researchers called it "software intervention". Software interventions can either positively or negatively affect the translation depending on whether the error improves readability without distorting meaning or inaccuracies. The study recommends the importance of tuning the AIPTSs applications to prevent the inducing of technical systems</dc:description>
	<dc:publisher xml:lang="en-US">Universitas Negeri Makassar</dc:publisher>
	<dc:contributor xml:lang="en-US"></dc:contributor>
	<dc:date>2025-01-01</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US"></dc:type>
	<dc:type xml:lang="en-US"></dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>http://ojs.unm.ac.id/ijole/article/view/70034</dc:identifier>
	<dc:identifier>10.26858/ijole.v8i4.70034</dc:identifier>
	<dc:source xml:lang="en-US">International Journal of Language Education; Vol. 8, No., 4, 2024</dc:source>
	<dc:source>2548-8465</dc:source>
	<dc:source>2548-8457</dc:source>
	<dc:source>10.26858/ijole.v8i4</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>http://ojs.unm.ac.id/ijole/article/view/70034/pdf</dc:relation>
	<dc:coverage xml:lang="en-US"></dc:coverage>
	<dc:coverage xml:lang="en-US"></dc:coverage>
	<dc:coverage xml:lang="en-US"></dc:coverage>
	<dc:rights xml:lang="en-US">Copyright (c) 2025 International Journal of Language Education</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/</dc:rights>
</oai_dc:dc>