Google Machine Translation - What are the examples and challenges in IT manuals?

In the previous blog, I wrote about Google Neural Machine Translation (GNMT) supporting Japanese.

If you want to know the meaning of an English website page personally, GMNT might be sufficient for a rough translation. However, when it comes to professional translation, the situation is different. When the target audience consists of product users or customers who have purchased services, a certain level of quality in translation is naturally required.

Many readers may have the question, "While I acknowledge that the quality of GNMT has improved, to what extent can its quality actually be evaluated?"

Therefore, we examined examples of translation quality and challenges in the main translation genres of Human Science: IT, medical, and mechanical documents. Please note that the GNMT translations are based on the time of writing this article.

IT Manual (English-Japanese)

The translation is correct in terms of meaning, but it is necessary to check whether it complies with the glossary and style guide.
For example, if the translation of "deployment" is standardized as "deployment" in the glossary, it needs to be corrected to "deployment" instead of "expansion".
Regarding style, it is necessary to correct points such as the presence or absence of spaces between full-width and half-width characters (between "API" and "of"), and whether to use full-width or half-width parentheses, according to the style guide.
Additionally, it is necessary to verify and correct whether the referenced section title is actually "Deployment Using Application Programming Interface (API)".

It is necessary to check whether the translation produced by Google Translate complies with the style guide. Additionally, since the example sentences are reference texts, it is also important to verify whether they match the section titles.
These manual verification processes are referred to as PE (Post-Editing).

Related Services

Neural Machine Translation Solutions

Machine Translation Seminar

Translation Open Seminar Scheduled