Machine translation vs.
translation memory
As part of an
initiative to implement a multilingual web site, companies should closely
evaluate machine-translation and translation-memory technologies. While these
technologies will not eliminate the need for the translation and localization
of web content, they are appropriate in some instances.
Following is a
brief overview of the two technologies and an assessment of their usefulness in
multilingual web-support systems.
Translation
Memory (TM)
TM software is
designed to enhance the human translation effort. The software stores matching
source and target language segments that were translated by translator in a
database for future reuse. As the translation effort progresses, the
translation memory grows. Newly encountered segments are compared to the
database content, and the resulting output (exact, fuzzy or no match) is
reviewed and completed by the translator.
Machine
Translation (MT)
MT software
aims to replace the human translator. The algorithm analyzes the grammar and
syntax of source segments according to previously defined rules. It then
queries a dictionary to produce a translated segment without human
intervention. MT output is generally not good enough to be published without
extensive human post-editing.
The primary
reasons for implementing either technology are speed, cost savings, and
consistency:
Depending on the languages, content volume, type
of source text and publishing timeframes, either one or both of these
technologies may improve the efficiency and reduce the cost of publishing
multilingual support content.
Andres
Heuberger is an editor at multilingualwebmaster.com. He frequently writes
on issues related to technology, translation, and regulations. Rants and raves
can be sent to aheuberger@multilingualwebmaster.com.
© 2001
ForeignExchange Translations