Machine translation and language professionals – If you just got your Japanese friend’s email in the spam it’s time to think about using MT

I have always been interested in language technologies. I was curious to discover the way technology could process natural language and help humans to communicate. However, I’ve always felt scepticism by a number of professionals about it. As you may imagine machine translation is one of the hottest subjects in Natural Language Processing: it rages the strongest reactions and not just among translators.

As an MT application was useful to me once, I think it’s time to support MT a bit more.

Let’s see why:

  • MT may be a cheap and quick solution to understand each other

If you work in the translation industry you get used to communicate with people coming from different countries and don’t realize your brain localize automatically. But even if you know some languages, there are a lot you don’t understand at all…Few days ago I received an email written in ideograms marked automatically as spam. I have a degree in languages, and guess what? I couldn’t understand a word, so I pushed the magic bottom “translate”. Once I machine translated it I found out that the email was a message sent by my former Japanese housemate. She sent it from a non-localized social network that’s why part of the message was in Japanese. Without an MT application I would have just deleted it.

  • MT works also with a lingua franca

The result I got was in German because my email is localized in this language. Even if it is not my mother tongue the MT output was useful to me. There is this debate in the NLP Industry according to which MT research should focus on few language combinations to cover faster a larger quantity of speakers. On the one hand if you use MT just to understand what the text is about this could be a good approach, on the other hand you risk focusing only on money-makers languages. Join the debate by having a look at this blog post about Nicholas Ostler’s book “The Last Lingua Franca”

  • MT won’t decrease overall translation quality.

Translation quality is different according to what it is done for. As online content is growing more and more, there is less and less time and budget for translating it. Have a look at this infografic published by SDL about how to survive in a digital content explosion world to get more information

An MT application was useful to me once and I guess I will be in the future, what do you think about it?


What is Finger Reader?

Finger Reader is a device that assists visually impaired users with reading texts or words. It’s
basically a ring the user wears on their index finger that houses a tiny camera and some haptic
actuators for feedback. When a visually impaired person wants to read some text around them,
maybe a business card, a menu in a restaurant, a sign, they point their finger at the surface with
the text and the device reads the words out loud. They can go faster, slower, go back, etc, and
get feedback if they are vearing off a line of text nudging their finger back in the right direction.

Finger Reader is just a prototype at this point what’s the next step?

Continue reading here

Rosetta Stone Buys Up Online Language Learning Community Livemocha For $8.5M In Cash

Originally posted on TechCrunch:

Rosetta Stonehas just acquired Seattle-based online language-learning community Livemocha for $8.5 million in cash. Livemocha has one of the largest online language-learning communities in the world, boasting over 16 million members from over 195 different countries.

Rosetta Stone will likely use Livemocha’s cloud platform to offer its language packages online — they were once only available via disc set, with a complimentary iPad app.

Here’s what Rosetta Stone CEO and President Steve Swad had to say about it:

We are in the process of transforming Rosetta Stone to be the most dynamic and ubiquitous technology-based learning platform in the world. Our acquisition of Livemocha will help accelerate that transformation. With Livemocha and its vibrant online community on our side, Rosetta Stone will reach more people and change more lives than ever before.

Livemocha will remain in its Seattle-based offices, adding yet another arm to Rosetta Stone’s US presence which…

View original 144 more words

Machine Translation in Short

Originally posted on Pangeanic Translation Technologies & News:

It is evident that certain documents require a human translator in order to interpret the subtleties of a language. Nevertheless, no matter how skilled a human translator may be, machine translation (also known as automatic translation or MT for short) exceeds the efficiency of a human translator.

Machine translation is generally used for subject-specific cases and this is where results and productivity rates are spectacularly higher. It allows individuals and companies to tailor their work according to the topic. Consequently, this enriches the output and quality of machine translation by cutting down on the number of choices for each word(s) to be translated.

This form of translation is extremely helpful in areas where formal language is used or phrases are repeated without much variation, such as administrative documents, which do not require the use of colloquial language and expression.

The potential of machine translation has been increasingly explored. In 2009…

View original 238 more words


Interview with Marja Vaba – Translation Quality Manager

Marja Vaba is a Translation Quality Manager working for one of the biggest and most innovative IT world companies. Her contribution is very useful to understand in which direction the translation/localization is going towards terminology management.

1) Linda Burchi: Could you please introduce yourself? What is your academic background, work experience and current position?
Marja Vaba: I have worked as Translation Quality Manager in a quickly growing software company for almost five years. When I started, it still felt like working in a start-up where going with the ideas was more important and beneficial than following documentation. Now the company is so big that following processes and having decent documentation and data-bases is essential for quality assurance in both software development and translation (quality) management. I have had the chance to build up the translation quality system in the company. I have found my background in linguistics and translation industry useful for that.

2) Linda Burchi: Is terminology an important variable on the web?
Marja Vaba: I think that terminology is an important variable in all texts because the consistent and systematic way we call things and phenomena is very important for reader/user understanding. There is also another aspect: in web, new products emerge very often, thus well-coined and motivated terms help users to understand better what they are buying/dealing with.

3) Linda Burchi: Has the web communication an influence on terminology?
Marja Vaba: It certainly does. In web, everybody has access not only to information, but also freedom to express their opinion. This also counts for terminology – kind of crowdsourcing is going on all the time if a company listens to its users/community.

continue reading here