Scribd
Upload
96 views17 pages

Computer Assisted Translation Tools

This document summarizes and compares the performance of four machine translation tools - Google Translate, Bing Translator, Yandex Translate, and SDL FreeTranslation - on a test involving compound words, idioms, missing words, and words with multiple meanings. It finds that Bing performed best overall, translating the missing word most clearly while also correctly translating the word with multiple meanings, though it translated the idiom literally like the other tools. Google performed the least well on compound words. Yandex scored full marks on compounds but had the worst translation of the missing word. SDL translated compounds correctly but produced incorrect sentences. All tools struggled with idioms.

Uploaded by

Mega
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
96 views17 pages

Computer Assisted Translation Tools

This document summarizes and compares the performance of four machine translation tools - Google Translate, Bing Translator, Yandex Translate, and SDL FreeTranslation - on a test involving compound words, idioms, missing words, and words with multiple meanings. It finds that Bing performed best overall, translating the missing word most clearly while also correctly translating the word with multiple meanings, though it translated the idiom literally like the other tools. Google performed the least well on compound words. Yandex scored full marks on compounds but had the worst translation of the missing word. SDL translated compounds correctly but produced incorrect sentences. All tools struggled with idioms.

Uploaded by

Mega
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

Computer Assisted

Translation Tools
SITTI NURJANNAH, S.PD. M.PD.
Learning Objective

In this chapter, we will explore the use of computer as a tool for


assisting translation process.
Computer-Assisted Translation

Computer-Assisted Translation is a term used to describe computer


techniques to facilitate the process of translation.

there are three basic types of CAT (Computer Assisted Translation) tools
to assist human translation work;
Machine Translation

• Machine Translation (MT) is an action of fully automatic text


translation.
Terminology Management

It refers to databases that hold preferred terms specific to an enterprise.


Essentially, a terminology database is an intel-ligent glossary of a
company’s preferred translations of specified terms in multiple
languages.
Translation Memory

• Translation Memory (TM) is a database of previously carried out


translations. TM contains only high-quality translations which can be
reused in future. The reuse of translations by means of Translation
Memory is the most widely recognized and appreciated feature of
CAT.
Potential Benefits of CAT

1. Repetitive or similar texts need be translated only once.


2. Once glossaries have been entered in the system, future translations will always
be consistent.
3. Greater speed of draft translation, thereby allowing more time for quality control.
4. A computer can work on draft translation at any time of the day, thus a 10,000
word translation that would take a human translator a week to produce could be
done overnight ready for editing the next morning. (Speeds of up to 10,000 words
per hour have been claimed).
5. Better quality control since text already entered in the software will not need to be
re- checked.
Machine Translation

• Machine translation (MT) is automated translation or “translation


carried out by a computer”, as defined in the Oxford English
dictionary. It is a process, sometimes referred to as Natural Language
Processing which uses a bilingual data set and other language assets to
build language and phrase models used to translate text.
Bing

• Bing had no problems at all with the compound words test. As with all
the tools we tested it translated the idiom ‘once in a blue moon’
literally. Bing did the best of all of the four tools with it’s translation of
the missing word ‘fundraising’ offering the most clear alternative. It’s
translation of the multiple meaning word ‘rose’ was correct, though
the sentence itself was wrong. Despite the error in the last sentence,
Bing’s translation was by a narrow margin the most understandable of
the four tools tested.
Google

• Google preformed the least well of the four tools on the test for
compound words, because it’s translation of ‘Signup’ although
understandable was not as good as the other three. Like all the tools
the idiom ‘once in a blue moon’ was translated literally. The
translation of the missing word ‘fundraising’ was good and there were
no problems with the multiple meaning word ‘rose’.
Yandex

• Yandex scored full marks on the compound word test, and like all of the
tools translated the idiom ‘once in a blue moon’ literally. It performed the
worst of the four with the translation of ‘fundraising’ though the
translation offered was just about understandable. Yandex had no
problems with the multiple meaning word ‘rose’. Like SDL it also seemed
to have a problem dealing with quotation marks, changing single quotation
marks into double but not closing them. We tested several sentences to
check this error and found it to be a common one. Sometimes double
quotes were used and not shut, at other times a combination of single and
double quotes were used.
SDL

• SDL translated all of the compound words in our paragraph correctly.


It translated the idiom ‘once in a blue moon’ literally like all the tools
we tested. SDL offered a good alternative for the missing word
‘fundraising’, and translated the multiple meaning word ‘rose’ with no
problems, though in both cases the sentences themselves were wrong.
As mentioned above SDL also seems to have a problem in dealing
with quotation marks.
Google, Bing Yandex, and SDL have different translation products.
Following are the samples of translation product by, Google, Bing
Yandex, and SDL. Can you tell the differences?
Google’s translation product:
https://translate.google.com
Bing’s translation product: https://www.bing.com/translator
Yandex’s translation product: https://translate.yandex.com
SDL’s translation product: https://www.freetranslation.com

You might also like