More and more language service providers in the media localisation industry are experimenting with and implementing machine translation (MT) solutions in their subtitling workflows. As I mentioned in an earlier blog post, it is important to think about the way in which MT integration is made in the user interface, and the features translators need so as to facilitate MT post-editing tasks.
Several studies have taken place over the years, in subtitling and other translation fields, to better understand the post-editing task and the impact it has on translators, their productivity and the quality of the translations they produce with the use of MT. Even though the samples are small overall, their findings are generally consistent and can be used to infer an initial list of requirements that any MT integration should cover if one is to achieve optimum results and make the most out of the technology.
I outline my thoughts on this topic with respect to implementations in the subtitling field in this article published by Multilingual.
Are you a subtitler? Have you used machine translation in your work? What is your experience so far? Is there any other feature you would like to see implemented?