Machine Translation Post-Editing (MTPE)

The translation industry is constantly evolving, and for quite some time we have been observing the rapid development of technologies supporting the translation process. Almost everyone has encountered machine translation. Until recently, the best results were achieved by Neural Machine Translation (NMT) engines, which in recent years have been combined with the capabilities of generative artificial intelligence and Large Language Models (LLMs). Regardless of the technology used, when opting for automated solutions, it is important to be aware of their limitations. Their optimal use is overseen by a human specialist – in this case, a post-editor. A post-editor is a translator who performs a specific type of verification known as post-editing.

What does post-editing involve?

Post-editing is a complex verification process that consists of reviewing the text in every respect, primarily through a detailed comparison of the source document and the translation, identifying any errors that appear in the translation, and ensuring terminological consistency throughout the text. Regardless of the MT engine used, the post-editor is responsible for the final quality of the document.

In practice, we distinguish between two main levels of post-editing:

  • Light post-editing focuses on basic corrections: removing errors in meaning, offensive wording, or culturally inappropriate elements. It does not involve improving style, grammar, or formatting.

  • Full post-editing requires significantly more work – it also covers grammatical, terminological, and punctuation accuracy, compliance with client guidelines, and preservation of formatting.

The choice of post-editing level depends on the intended use of the text. The light version may be sufficient for internal document circulation and for materials translated primarily for content comprehension and internal use. For texts intended for external use, marketing materials, or specialist content, full post-editing is essential.

What are the advantages and disadvantages of machine translation with post-editing?

The two main advantages of using machine translation with post-editing are:

  • reduced turnaround time (especially for very large projects), and

  • cost savings.

However, it is important to remember that not every text is suitable for machine translation and, consequently, for post-editing (e.g., literary or marketing texts). One of the fundamental challenges of machine translation is the lack of terminological consistency and the inability to properly interpret cultural context.

In the case of machine translation with post-editing provided by BTD Services, the result is always a high-quality document that meets stringent quality standards.

Ask for a free quote