Translation management system like XTRF can largely streamline the process of handling translation projects and increase the working efficiency of project managers so that they can mange more projects.

 

In this article, you will see how exactly XTRF is helping project mangers as well as vendors and clients in the whole process. And this article will also talk about the improvements and shortcomings of this website. Hope you could have a clearer understanding of what translation management system (or XTRF) is after you finish reading this article!

 

Here are several advantages and disadvantages of XTRF:

 

• PM Portal, Client Portal and Vendor Portal

 

XTRF has three different portals: PM Portal, Client Portal and Vendor Portal.

 

[PM portal]

 

[Vendor portal]

 

[Client portal]

 

As you can see from the screen shots above, each portal has a different color and a different format so that people can easily find out which portal they are using.

 

 

• Saving time of creating quotes and POs

XTRF can save PM time in creating quotes and POs. In the XTRF website, it can automatically generate the quotes and PO in a uniformed format, which you can see from the example screen shot below. XTRF can largely increase the efficiency for PMs when creating quotes and POs.

[Quote]

 

• Checking vendor availability and assigning works to vendor more easily

[Vendor]

 

In the PM portal, you can see a list of vendors, their overall evaluation, languages combination and other contacting information. You can either assign a job to a specific person by searching his/her name in the “Text” box at the top left of the page, or you can assign the work to a range of vendors and set the rule that the person who accept the work first can get the job.

 

By doing this, you do not need to spend a lot of time checking your contact book and call each of your vendor and check their availability. XTRF will automatically send them an email announcing them this job, and they can accept it by clicking the link provided in the email.

 

 

• Preparing and sending files

[Managing projects]

 

As you can see in the PM portal page, PM can manage different projects at the same time. Without the help of XTRF, it is very possible that they will miss a document or send the wrong document to the wrong person. However, by using XTRF, you can upload the file you want the vendor to translate on the platform. After the vendor translated the document, he/she can upload the translated document on the platform so that the proofreader can start his/her job as soon as the vendor upload the document.

 

• Better transparency for clients

Using XTRF can help the client know the progress of the whole project and create a sense of transparency for them.

 

• In case of emergency…

As a PM, there will be some time when you need to help your client or your vendor do their job. For example, your client told you he/she cannot access to the Internet or he/she knows nothing about XTRF. In order to save more time and help everything go smoothly, you can sign in the vendor portal and client portal and help them do their works.

 

However, it might be confusing and difficult for the PM to log in as a vendor or client because the “secret path” might be difficult to find.

[sign in as the vendor]

[sign in as the client]

 

• Price and competitors

 

XTRF is very powerful, but it is also not cheap.

[XTRF price]

 

As you can see from the screen shot above, it costs $129 per user per month.

 

In the market, there are other similar translation project manager systems like SDL TMS and Smartling Translation Management System. The latter is free for translators and agencies.

 

[competitors]

So for smaller translation agencies, although XTRF can largely improve the working efficiency, spending so much money on platforms such as XTRF may be a tough choice.

 

 

That is how you saw in the Google translation video: you went to a country where you cannot speak their language. You grab your phone and open the Google Translate. So now you can just be as anybody else. Language barrier is no longer there. You can use Google Translate to translate everything you see, hear for you.

Google translation is so powerful. You can get instant translation when you hold your camera to a source language text. You can speak to the microphone in your own language and get the translated words in a few second.

 

As we have learned in the Intro to Computer-Assisted-Translation class, Google Translate is a SMT (Statistical Machine Translation), which means translations are generated on the basis of statistical models. All the translations in Google are based on tons of millions of translated documents.

 

As a former linguistic major student, I have learned in my Semantics and Pragmatics class that one of the main differences between human beings and animals is that human beings have the complicated system of grammar. For many years, scientists have tested on many smart animals such as gorillas and dolphins to teach them build up the awareness of grammar but all of them have failed. Since human being is the only creature in the world that know how grammar works, would it be possible that machine could have the same brain as human beings? Can machine be as good as human brain?

 

• Here are some major problems I found out in Google Translate:

 

• Unable to translate/pronounce tones

For Google Translate, you can speak to the microphone and the machine will automatically transcript what you have said into words. After the words have been translated, it is then spoken by the machine again. This leads to two problems:

1) the tone in the source sentence cannot be captured;

2) the original tone cannot be integrated in the translated sentence and then cannot be heard by the listener.

In this example, the Chinese sentence “你真的可以吗” is a rhetorical question, which means “Do you think you can do this?”. However, the translation is totally wrong. First, it ignored the exclamation word “吗”, and then literally translated each word of “你真的可以” into “You can really do”, which does not make any sense in English. Also, when hearing the translation, it is a declarative sentence.

In this example, the Chinese input sentence “好玩吗” was a rhetorical question which means “Is it fun?”. Google translated into “Is it fun”. This may seem right at the first sight. However, when the machine speaks this sentence, it is pronounced into a declarative sentence and made the tone went down at the end of the sentence. This gives listeners a feeling that this sentence is a sarcastic sentence, which means: “Someone wants to make something fun. But actually it is very boring.”

 

However, in some cases, the machine can detect the exclamation word and make changes in the translation.

In the first example, Google Translate successfully detected the exclamation word “吗” so it translated into “Are you serious” which is a rhetorical question. And in the second one, it translated into a declarative sentence “You are serious”.

 

SMTs like Google Translate need to use more and more data to make the program itself perfect.

 

•  Ambiguity

In spoken Chinese, people sometimes say broken sentences that can make sense in a certain context but they are not necessarily grammatically correct. This may be difficult for machine translations to translate because it is very likely that the machine will translate it wrong.

In this example, “你说”
can mean “What do you think?” “What you said is…” “I will let you decide.” and so on. However, the machine translated it into “You said”, which may not be appropriate in that context.

 

When encountering this problem, translation machine can provide people with different translations so that the listener could select the one that is the most appropriate.

 

• Conclusion:

Using a real time machine translation to communicate with a foreigner who speaks a totally different language may be possible, but sometimes it may cause problems because of the imperfection of machine translation. Unable to detect and pronounce the exclamation word and tone may cause the wrong translation.