Post-editing machine translated subtitles: examining the effects of non-verbal input on student translators’ effort

Despite the potential productivity benefits of post-editing machine-translated outputs, post-editing practices have rarely been integrated into the workflow of subtitle production. This study examines the usefulness of employing post-editing methods during the subtitle production process by comparing student translators’ translations to post-editing machine-translated subtitles. The amount of non-verbal input originating from the provided audiovisual materials was also included as an independent variable to identify its potential effect on the subtitle post-editing process. An experiment was then conducted that utilized an English-Chinese language pair and eye-tracking and keystroke-logging techniques to estimate student translators’ effort during both translation and post-editing processes. The results of the study reveal that there is a significantly shorter production duration and fewer keystrokes typed during the post-editing process, with texts that contained increased non-verbal input proving to be significantly less cognitively demanding as evidenced by this study’s observations of the fixation count and the overall duration.