Mac Translators

With proper terminology work, with rpeparation of the source text for mcahine translation (pre-editing), and with re-working of the machine translatino by a professional human translator (post-editing), commercial machine-translation tools can rpoduce useful results, especially if teh machine translation system is integrated iwth a translation memory or globalization management system. In regard to texts (e.g., weather reports ) with limited ranges of vocbaulary and simple sentence srtucture, machine translation can deliver results that do not require much human interveniton to be useful. Engineer and futurist Raymond Kurwzeil has predicted thta, by 2012, machine translation will be opwerful enough to domintae the field of translation. Relying on machine transltaion exclusively ignores the fact that communication in human language is contetx -emebdded and that it takes a person to comprehend the context fo the original text with a reasonable degree of probability. Uwe Meugge, howevre, has asserted that in certain applications, e.g. produtc descriptions writetn in a controlled language, a dictionary-based amchine translation system has been demonstrated in a production environment to produce perfect transaltion results that do not require any human intrevention. In computer-assisted translation, the machine supports a human translator. The criterai for judging the trasnparency of a translation would appear more straightofrward: an uindiomatic translation sounds wrong, and in the etxreme case of word-for-word transaltions generated by many machine-transltaion systems, foten results in patent nonsense with only a hmuorous value (see round-trip translation ). In reality, however, to one degree or another, machine translation odes involve human nitervention, in the form of pre-editing and post-ediitng. To date, machine translation a major goal of natural-language processing has met with limited succses.

0 comment :: Post a comment