ACCEPTABILITY OF MACHINE TRANSLATION IN TRANSLATING HUMOR OF PAUL SNYDER STAND-UP COMEDY

REYDINA ISNATA CHURULLIAIN, 1860203222092 and NANY SOENGKONO MADAYANI, 197305152007102003 (2026) ACCEPTABILITY OF MACHINE TRANSLATION IN TRANSLATING HUMOR OF PAUL SNYDER STAND-UP COMEDY. [ Skripsi ]

	Text COVER.pdf Download (609kB)
	Text ABSTRACT.pdf Download (77kB)
	Text Table of Content.pdf Download (162kB)
	Text CHAPTER I.pdf Download (235kB)
	Text CHAPTER II.pdf Restricted to Registered users only Download (279kB)
	Text CHAPTER III.pdf Restricted to Registered users only Download (239kB)
	Text CHAPTER IV.pdf Restricted to Registered users only Download (402kB)
	Text CHAPTER V.pdf Restricted to Registered users only Download (182kB)
	Text CHAPTER VI.pdf Restricted to Registered users only Download (144kB)
	Text REFERENCES.pdf Download (177kB)
	Text APPENDICES.pdf Restricted to Repository staff only Download (945kB)

Abstract

The thesis entitled "Acceptability of Machine Translation in Translating Humor of Paul Snyder Stand-Up Comedy" was written by Reydina Isnata Churulliain, Student ID Number 1860203222092, English Education Department, Faculty of Education and Teacher Training, Universitas Islam Negeri Sayyid Ali Rahmatullah Tulungagung, supervised by Nany Soengkono Madayani, SS., M. Pd. Keywords: Acceptability, Humor Translation, Machine Translation, Stand-Up Comedy, Google Translate, DeepL, ChatGPT The rapid advancement of machine translation (MT) technology has expanded its application to various types of text, including creative and culture bound content such as humor. Translating humor is widely regarded as one of the most demanding tasks in translation, as it requires not only linguistic accuracy but also sensitivity to cultural references, informal register, and pragmatic nuance. Despite the growing use of MT tools in everyday communication, relatively little research has examined their performance specifically in the context of humor translation, particularly for the English-to-Indonesian language pair. This study uses 20 humorous utterances drawn from Paul Snyder's stand-up comedy special as its primary data source. The study addresses two research questions: (1) how acceptable are the translations of humor produced by three MT tools from English into Indonesian; and (2) which MT tool performs best overall in preserving the intended meaning and comedic effect of the source text. The objective of this study is to evaluate and compare the acceptability of translations produced by each tool, and to identify the linguistic and cultural factors that most significantly affect translation quality in the context of spoken humor. This study employs a qualitative descriptive approach. First, the full video was transcribed verbatim, and 20 humorous utterances were identified using Attardo and Raskin's General Theory of Verbal Humor (GTVH). Each utterance was then translated by Google Translate, DeepL, and ChatGPT. Three trained raters evaluated the translations using the acceptability rubric adapted from Nababan, Nuraeni, and Sumardiono (2012). Finally, a Focus Group Discussion (FGD) was held among the raters to consolidate their scores. The findings show that Google Translate scored 1.42 (Unacceptable), DeepL scored 1.58 (Unacceptable), and ChatGPT scored 2.35 (Acceptable). The main factors affecting acceptability were formal register, untransferred cultural references, and inconsistent pronoun use. Despite minor limitations, ChatGPT produced the most natural and colloquial translations overall. This study concludes that ChatGPT is the most suitable MT tool for translating humor from English stand-up comedy into Indonesian, while Google Translate and DeepL still require human post-editing for this type of content.

Actions (login required)

View Item