TT 2 - TransType 2 ( IST - IST - 2001 - 32091 )

MAHT systems do not attempt to produce target texts without the intervention of the human translator but to help him/her in achieving highe r productivity rates while increasing the final quality. The first goal is commonly achieved by building a translation memory, a (usually large) base of source-totarget text fragments that the system tries to employ when newly (though similar) source text is presented to the system. Consistency, at the sentence level, is ensured by reusing previous (approved) translations; at the terminological level, MAHT tools help translators to build and share terminological databases. Terminology management tools are included almost in every application that attempts to automate the translation process, particularly in MAHT applications but also in MT systems. The TT2 system could be considered as a CAT tool, in the broad sense of the term, since it is conceived as a translation aid that attempts to increase productivity and quality in translation. However, it departs from current CAT tools in the technology employed for achieving those goals. Commercial CAT tools employ previous translations (translation memories) for suggesting new translations and these translations are exploited via string-matching procedures (either exact or fuzzy). In contrast, TT2 exploits previous translations using statistical MT techniques; as a result TT2 is not tied to exact or fuzzy matches at the sentence level. TT2 can be described as an interactive machine translation (IMT) system.