Columns related with “Localization and translation”
How to use translation support software and machine translation
Since the second half of the 2000s, computer-assisted translation software such as SDL Trados has become firmly established in fields such as industrial translation.
In addition, since the end of the 2000s, the performance of machine translation has improved dramatically, and it is now practical if used appropriately.
Merits of translation support software
Trados and similar translation support software assist the translation process by automatically referencing and applying existing translations stored in a translation memory to new translations based on their similarity.
The software aims to reduce the amount of translation and increase work efficiency by accurately reusing existing translations, so it is effective for creating documents that have many common or similar parts, such as instruction manuals. Since the same translation will be used wherever possible for the same wording, using it with terminology management software such as MultiTerm will also help to standardize wording and phrasing.
In addition, if the document is in a compatible file format, the translated file can be output with the original text automatically replaced by the translation. It reduces the costs of DTP and HTML editing.
Utilizing machine translation
Machine translation services such as Google Translate and DeepL have dramatically improved their translation accuracy in recent years, and are now able to produce very natural translations. On the other hand, however, it is common to see entire sentences not being translated or negative sentences being turned into positive ones, so currently it is essential that a human check is done to ensure that the translation is appropriate.
When using machine translation beyond its intended use as a computer-assisted translation tool for translators or a quick check on completed translations, it is necessary to create a solid system for checking the validity of the translations.