PENERJEMAHAN B.INGGRIS BERBANTUAN KOMPUTER (2 assignment)

4SA04, GROUP 2

Members of the Group:

  1. Annisa Sutan Zainal (10612981)
  2. Aryani (18612122)
  3. Melania Santosa (1461253)
  4. Romy Chandra (16612680)
  5. Tiya Rahmawati (17612418)

 

Advantages Of Using SDL Machine Translation

SDL is the leader in language translation and global content management.  With more than 20 years of experience, SDL helps companies build relevant digital experiences that deliver transformative business results on a global scale.

 

Bespoke, high-quality output

Complete your projects faster than ever, while preserving translation quality, with the ability to train your own machine translation engines. SDL Language Cloud Custom MT Engines provide translators with a source of bespoke, high-quality MT output that requires minimal post-editing before publishing.

Boost translation productivity

Achieve new levels of productivity by tailoring your own engines to your specific requirements; specific clients, projects or industry verticals. A trained engine can translate industry terminology 24% more consistently than an untrained engine, helping you work more efficiently.

The benefits of Custom MT Engines

  • Trainable– Enhance translation quality with the ability to train your own engines
  • Secure – Your data is protected and translated over a secure, encrypted connection
  • Cloud-based– No hardware costs, easy to get started, no maintenance
  • 100 baseline language pairs– Try various language pairs quickly and easily with no additional costs
  • Seamless – Designed to integrate with SDL’s world leading language technologies
  • Flexible pricing – A flexible package that allows you to use machine translation as often as you wish

SDL Translate is your ‘all in one’ translation solution. It’s a communication platform, phrase dictionary, language teacher, and travel companion. With over 90 language pairings including Spanish, German, French, Italian, Chinese and Arabic (www.freetranslation.com), SDL Translate covers every corner of the globe.

Key Features:

– Text to speech (TTS) * : Write a sentence and tap the translation to hear it.

– Speech to text (STT) * : Tap the microphone button and speak to get your sentence translated.

– Phrases: Introductory Phrases in Spanish, German, and French. Available without a network connection.

– Idiomatic Expressions: Ever wanted to know how to say “Piece of Cake” in another language? Now you can translate 65 Idiomatic expressions from English to Spanish, French, or German.

* This feature depends on the available languages on your device.

The following languages are available for Translation:

Western Europe:

Danish ⇨ English
Dutch ⇨ English
English ⇨ Albanian
English ⇨ Arabic
English ⇨ Bengali
English ⇨ Bulgarian
English ⇨ Chinese (Simplified)
English ⇨ Chinese (Traditional)
English ⇨ Czech
English ⇨ Danish
English ⇨ Dutch
English ⇨ Estonian
English ⇨ Finnish
English ⇨ French
English ⇨ German
English ⇨ Greek
English ⇨ Pashto
English ⇨ Hausa
English ⇨ Hebrew
English ⇨ Hindi
English ⇨ Hungarian
English ⇨ Indonesian
English ⇨ Italian
English ⇨ Japanese
English ⇨ Korean
English ⇨ Latvian
English ⇨ Lithuanian
English ⇨ Malay
English ⇨ Norwegian
English ⇨ Polish
English ⇨ Portuguese
English ⇨ Persian
English ⇨ Romanian
English ⇨ Russian
English ⇨ Spanish
English ⇨ Somali
English ⇨ Serbian
English ⇨ Slovak
English ⇨ Slovenian
English ⇨ Swedish
English ⇨ Thai
English ⇨ Turkish
English ⇨ Ukrainian
English ⇨ Urdu
English ⇨ Vietnamese
Finnish ⇨ English
French ⇨ English
French ⇨ Arabic
French ⇨ Spanish
French ⇨ German
Spanish ⇨ Italian
German ⇨ English
German ⇨ French
German ⇨ Spanish
Greek ⇨ English
Italian ⇨ English
Italian ⇨ Spanish
Norwegian ⇨ English
Portuguese ⇨ English
Spanish ⇨ English
Spanish ⇨ Arabic
Spanish ⇨ French
Spanish ⇨ German
Swedish ⇨ English

 


Eastern Europe:

 

Albanian ⇨ English
Bulgarian ⇨ English
Czech ⇨ English
Estonian ⇨ English
Hungarian ⇨ English
Latvian ⇨ English
Lithuanian ⇨ English
Polish ⇨ English
Romanian ⇨ English
Russian ⇨ English
Slovak ⇨ English
Serbian ⇨ English
Slovenian ⇨ English
Ukrainian ⇨ English

Middle East & Africa:

Arabic ⇨ English
Arabic ⇨ Spanish
Arabic ⇨ French
Hausa ⇨ English
Hebrew ⇨ English

Pashto ⇨ English
Persian ⇨ English
Somali ⇨ English
Turkish ⇨ English

Asia:
Bengali ⇨ English
Hindi ⇨ English
Korean ⇨ English
Urdu ⇨ English
Chinese (Simplified) ⇨ English
Chinese (Traditional) ⇨ English
Indonesian ⇨ English
Malay ⇨ English
Vietnamese ⇨ English
Japanese ⇨ English
Thai ⇨ English

Disadvantages Of Using SDL Machine Translation

There are some disadvantages when we use SDL as an alternative machine translation to help us during translating the text, there are:

  • When we translate a text use SDL Machine Translation, the result of translation isn’t natural. It depends on the original text. When the translator tries to quote a sentence or idiom or paraphrase from a novel; nevertheless, SDL Machine Translation translates the passage unnaturally.
  • The accuracy isn’t fully offered by SDL Machine Translation on a consistent basis. Usually, we can get the gist of the texts or anything else like that, but SDL Machine Translation only does word to word translation without comprehending the information which might have to be corrected manually later on. SDL Machine Translation rarely reaches accuracy levels above 70%, while a human translation almost always produces accuracy above 95%.
  • SDL Machine Translation is based on formal and systematic rules, the inferior translation quality of the texts with ambiguous words and sentences. So, sometimes SDL Machine Translation cannot solve the ambiguity by concentrating on a context and using experience or mental outlook as a human translator.
  • SDL cannot translate phrase, idiom, countable and uncountable noun correctly. It can be proved in the next page of this assignment.

 

The translation result of SDL

Due to the size of SDL application is very big, it is about more than 300mb, we translate the text with the website of SDL https://www.freetranslation.com/

  1. Translating Noun Phrase with SDL

sdl 1

From the screenshot above, it shows that in translating Indonesian noun phrase wanita muda yang sangat cantik, the result in English is the young woman was very beautiful. If we look at the result of the translation, the structure is grammatically correct, but the tense it should be simple present tense not simple past tense. In source language the form is a noun phrase, meanwhile in the target language the form changes, become a sentence.

Comparing to Ginger

ginger 2

If we compare SDL to Ginger, Ginger produces a better result than SDL. Indonesian noun phrase wanita muda yang sangat cantik is translated into a pretty young lady, in source language the form is a noun phrase, meanwhile in the source text the form is translated also into a noun phrase.

2. Translating Idiom with SDL

sdl 3

sdl 4

From the screenshot above, it proves that SDL cannot translate idiom correctly. In the first picture, we have example in source language English idiom break a leg which is translated into Indonesia becomes mematahkan kaki, SDL translates the idiom literally. English idiom break a leg has the same meaning with Good luck!. Second picture also proves that SDL cannot translate idiom correctly, idiom such give it a shot is translated into memberikan suara tembakan in Indonesian target language, the meaning of idiom give a shot is make an attempt or effort to do something, or in Indonesian it can be translated into mencobanya.

3. Translating Countable and Uncountable Noun with SDL

sdl 2

From the screenshot above, it shows that in translating countable and uncountable noun using SDL from Indonesian source language kami membeli banyak sekali keju dan bunga into English target language we buy a lot of cheese and interest. In the target language the structure is grammatically correct, and it uses simple present tense. The quantifier a lot of can be used for both countable and uncountable noun, so the use of a lot of is correct. The word cheese is uncountable noun, we cannot add suffix –s, SDL translates it correctly. Unfortunately the source language bunga is translated into interest in target language, it should be translated into flowers, it proves that SDL is not able to recognize context of the text.

 

Comparing to Ginger

ginger 3

Ginger translates the Indonesian source language kami membeli banyak sekali keju dan bunga into English target language we bought lots of cheese and flowers. The structure is grammatically correct, it uses simple past tense, which is correct. Unlike the result produced by SDL, ginger translates banyak sekali into lots of, it is correct, because quantifier lots of is used for both countable and uncountable noun. Keju dan bunga is translated into cheese and flowers, ginger translates it correctly, we cannot add suffix –s in plural form of the word cheese, because it is uncountable noun. Unlike SDL, translating bunga into interest, ginger translates it into flowers, which is correct. We add suffix –s in plural form of countable noun, flower is countable noun, so we add suffix –s becomes flowers. Ginger is able to recognize the context of the text.

We can take a conclusion, SDL doesn’t produce a good translation, ginger translates better than SDL.

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s