• Название:

    Computer Aided Translation Lecture 6

  • Размер: 0.03 Мб
  • Формат: RTF
  • Сообщить о нарушении / Abuse

    Осталось ждать: 20 сек.

Установите безопасный браузер

Предпросмотр документа

Computer-Aided Translation

Computer-Aided Translation (CAT) – is translation of texts with the help of computer technologies. It should not be confused with Machine Translation (MT), as the process of CAT is done by man; computers are used only to help translate a text either in less time or with better quality. The idea of CAT appeared with the computer itself: translators have always deprecated the generally accepted at that time conception of MT, which was profoundly researched in the branch of computer linguistics. At the same time, they supported the idea of using computers to help the translation. In 1960th, the European Coal and Steel Community (forerunner of the European Union) began creating terminology databases with the general title Eurodicautom. In order to create such databases, the all-Russian (ex All-USSR) Institute for Scientific and Technical information of RAS (ex USSR AS) was founded. (RAS – Russian Academy of Sciences)

The CAT modern concept had been developed in Martin’s Kay work (1980), where he pointed out such a thesis: taking over what is mechanical and routine, it (computer) frees human beings for what is essent ially human.

Today, the most spread ways of using computers for translation are: work with dictionaries and glossaries, Translation Memory (TM), which contains examples of texts that had been translated earlier. Also using of so-called package, large text collections in one or several languages, that give short descriptions of using words and phrases in general language or some specific subject field.

Different specialized tools are often used for software localization, for example, Passolo. They make it possible to translate menus and messages in software resources and programs which have been compiled already, and also to validate the localization. In order to translate the audiovisual materials (predominantly movies) specialized tools are also used (for example, Swift). They contain some aspects of translation memory; though additionally can provide the option of subtitles appearing in proper time, their format, conformance to video standards etc. Using CAT tools in simultaneous translation is very limited. It can be compared with using dictionaries, downloaded to a PDA. Another example is semiautomatic picking up of terminology lists while getting prepared to simultaneous translation in some specific field.

Having a large number of source texts and set terminology in specific fields; translators can also use the machine translation to provide good quality terminology and set phrases translation in a specific field. In this case, the translator makes post-editing of target text. Today considerable quantity of texts (mainly correspondence and legal texts) is translated with the help of CAT.

CAT Tools

Computer – Aided Translation is a broad and imprecise concept, which comprises a vast spectrum of simple and compound tools. For example:

·Spelling Check software, that can be integrated into text editors or additional programs;

·Grammar Check software, that is also integrated into tex t editors or additional programs;

·Terminology Management software, which helps the translator to manage his own terminology database in electronic form. That can be either a simple table, created in text editor, or electronic table, or database, created in FileMaker Pro program. For more laborious (and more expensive) tasks other software is available, for example LogiTerm, MultiTerm, SDL Trados, Termex etc;

·Mono- and multilingual vocabularies on CDs;

·Terminology databases, stored on CDs or the ones which connect to the Internet, for example Open Terminology Forum or TERMIUM;

·Full Text Search Software (or indicators), which helps to make a request to already translated texts or different reference material. Such indicators as Naturel, ISYS Search Software and dtSearch are well known in translation industry;

·Concordance Software, that helps to find the examples of words and phrases used in mono-, bi-, or multilingual package, for example bitext or translation memory;

·Bitext is a recent novation, that is a result of merging original text and its translation, which can later be analyzed by Full Text Search or Concordance programs;

·Project Management Software, which helps linguists to structure complex translation projects, transfer different tasks to different people and supervise the implementation;

·Software for Translation Memory Management (TMM), which comprises of text segment databases in original languages and their translation in one or several languages;

·Almost fully automatic systems, which resemble machine translation, though it is permitted that the user can make alterations in dubious cases. Sometimes such software is called human assisted machine translation;