Post-édition et traduction humaine en contexte académique une étude empirique
Main Article Content
Abstract
Today more than ever, translators must acquire new skills if they want to keep abreast of recent and future technological developments. The demand for machine translation (MT) post-editing (PE) has been growing steadily since 2017 and most PE work is being done by translators. Neural MT advances are clearly undeniable, but how does PE quality compare to that of human translation (HT)? To answer this question, we conducted a controlled experiment with translation students for the language pair English-French. In this paper, we present the results of the comparative analysis of our corpus after it was assessed by human evaluators. Our results showed, in particular, a higher success rate in PE of neural MT compared to both other methods (HT and PE of statistical MT), as well as a leveling effect in PE of neural MT.
Keywords: ¨Post-editing, Neural Machine Translation, Controlled Experiment, Translation Quality Assessment, Translation Students.
Downloads
Article Details
Proposed Policy for Journals That Offer Open Access
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).