RT info:eu-repo/semantics/article
T1 Google Translate vs. ChatGPT: Can non-language professionals trust them for specialized translation?
A1 Sanz Valdivieso, Lucia
A1 López Arroyo, María Belén
K1 5701.12 Traducción
AB Experts and professionals in specialized fields often need writing toolsto communicate in English as a means to disseminate their knowledge or enterthe international market. There are different tools to accomplish this and most ofthem are, lately, Machine Translation systems (MT) based on Neural MachineTranslation (NMT), an approach using artificial neural networks to translate withoutstanding fluency. Free and open systems such as Google Translate or, morerecently, ChatGPT used as a translator, have popularized NMT to a multitude ofusers. However, there are experts and professionals who, due to their lack ofcommand of English, often fail in their communication tasks by accepting NMTsystem’s output as correct. This paper examines these systems’ performancewhen translating terminology of the discourse in wine and olive oil tasting notes,specifically from Spanish into English. This domain may serve to represent lessstudiedspecialized languages where general language words and terms becomeclosely intertwined. The aim is to determine whether these systems can translateterminology accurately within the domain, and, if so, whether the GPT-3.5 modeloutperforms Google Translate. Results will help identify or discard possible languagesolutions for users who need to obtain texts in specialized English withprofessional and internationalization purposes, but who do not have the linguisticor economic resources to ensure the quality of the English text. Results show that,although ChatGPT yields fewer terminological errors than Google Translate interms of error severity and number of samples affected, professionals cannot relysolely on these tools just yet.
PB Hit IT
YR 2023
FD 2023
LK https://uvadoc.uva.es/handle/10324/84393
UL https://uvadoc.uva.es/handle/10324/84393
LA eng
NO Proceedings of the International Conference HiT-IT 2023
DS UVaDOC
RD 16-may-2026