With the rapid increase in the use of Machine Translation (MT), MT Post-editing has emerged as a growing need in the language industry. Since translation software is constantly being developed and improved, more and more companies are choosing to use MT for their projects.
MT is a process performed by a software piece whereby a written text in one language is rendered into another language without human intervention. There are three main approaches to MT (you can check our article “The Importance of Machine Translation in Today’s World,” where we explain each of them in detail):
The fact that translations rendered by MT are becoming increasingly accurate doesn’t mean that translators’ jobs are at risk. Indeed, MT Post-editing is still required to deliver quality translations.
Machine Translation Post-Editing (MTPE) is the process of reviewing, editing and correcting a translation rendered by a piece of software to achieve the required quality for translation. There are two main types of MTPE:
Despite the advantages mentioned above, MT Post-editing can also present some difficulties. For instance, depending on the client’s demands, a second round of revision (post-editing + editing) carried out by another translator might be required. Moreover, junior translators may not have enough experience to efficiently conduct a post-editing process and, consequently, the use of a more experienced translator may be needed.
Generally, post-editing rates represent 60% to 75% of the full per-word translation rate. However, post-editors’ work can vary (as mentioned before, they can make either minimal or more detailed corrections, depending on the scope of the task), and applying a uniform rate for all post-editing work doesn’t reflect this difference.
A way to solve this issue is by resorting to dynamic MT Post-Editing pricing, which measures the amount of editing (percentage) required by each MT segment in order to achieve the desired level of quality. The following picture shows this measurement: