AI in the crossbase translation process

Böblingen, 22.07.2022: The crossbase translation process determines the translation requirements in all necessary foreign languages and manages the translation jobs.

The established way: standardized text and translation workflow

The management of text modules as well as a sophisticated text workflow allow only missing or outdated translations to be fed into the translation process. This leads to a minimization of the amount of text to be processed. In addition, crossbase takes into account texts that are in translation and reacts to changes in the original text, for example with a warning.

crossbase has developed a tool for this purpose in order to be able to select different contexts for which a translation is needed (e.g. websites, e-catalogs, structures, print publications, etc.).  Thus, translation is only done where there is a need.

Pre-translation with DeepL

For more than three years, DeepL has been directly integrated as a machine translation solution. Individual or combined texts can be machine translated at the touch of a button. In a second phase, the texts can be assigned to the target country for review. Editing is done easily and conveniently via the crossbase web interface.

The dilemma with machine translation is that customer-specific dictionaries often cannot be used. This leads to extensive correction loops, especially for companies that use very specialized terminologies. In addition, it is very tedious to correct the same terms over and over again. 

NEW: The integration of glossaries for DeepL

Starting with version 508, glossaries were integrated into the machine translation in crossbase. This allows users to specify source and target language glossaries through the crossbase translation tool. The technical terms are thus translated correctly and the effort for correction is significantly reduced.

NEW: Connection to Microsoft Translator

Microsoft Translator can additionally be used for cloud-based machine translation. This offers more languages, see https://azure.microsoft.com/en-us/services/cognitive-services/translator/. This Azure service is very secure, scalable and offers a dictionary feature as well.