Machine translation and post-editing

We want to provide our customers with solutions for different translation needs, also large-scale, structured translation masses. Such text masses may be suitable for machine translation, provided that they are non-fiction texts with significant amounts of repetition and a regulated structure. In these cases, there is usually such a large amount of text to be translated that conventional translation may be slow and not very cost-efficient.

There are many different machine translation systems. However, today, the most interesting method is neural machine translation (NMT). Artificial neural networks process entire sentences in their context and the quality of the resulting translations is improving at a fast rate.

LanguageWire has a dedicated team of machine learning experts who are constantly developing our solutions. Our neural machine translation engines are efficient and highly secure: You can rest assured that translation always takes place in a secure environment and the customer’s materials do not leak into public channels.

Machine translation usually needs to be accompanied by post-editing. This means that a human translator or editor checks the machine-translated text, corrects any errors and edits clumsy structures. When a machine translation system is trained to produce text that meets the defined criteria, the combination of machine translation and post-editing may generate significant cost savings and efficiency improvements.