Merlin: Translation News, Services & Directory

This is the last post on this blog. The blog has moved to http://blog.gts-translation.com.

Localization World 2009, one of the main localization industry conferences which took place in Berlin this week was an impressive tour de force which proved how far the industry has advanced in recent years. Buyers from some of the largest corporations in the world were represented including Microsoft, SAP, Siemens, Google, Sun, Adobe and others. Machine Translation (MT) technology, long perceived as being too unreliable for mainstream commercial use, is now coming of age and seemed to stand out as one of the central themes in the conference workshops and exhibit booths.

But the most dominant issue of the conference was the worldwide recession and how it is impacting the industry. In his brilliant keynote speech on the first day of the conference, Deutsche Bank chief Dr. Norbert Walter analyzed the world events leading to today’s dismal economic conditions. Dr. Walter stated that this is worst recession any of us have ever witnessed in our lifetimes. Dr. Walter also provided some political predictions such as the winner of the next general election in German (Merkel) and said that US President Obama will lead the world out of the economic crisis.

The second day keynote panel also focused on the recession and how it is impacting the localization industry. Here too, machine translation technology dominated the panel as a possible way for the localization industry to revitalize in these tough times by enabling companies to get more for less. Francis Tsang, director of globalization at Adobe Systems Inc., opined that indeed the recession is a powerful driver behind the advance of MT and is probably one of the only good things that will emerge from the economic slowdown.

The Google Translator Toolkit which was released on June 9, 2009 is the latest in the line of free translation software products Google started to introduce a few years ago. The toolkit provides a suite of tools designed to assist translators in post-editing of machine translated content.

Post-editing of MT content has been a hot topic in recent months. The shortcomings and inaccuracies of MT are well known to all. Post-editing of MT by competent translators is perceived as a way of getting reasonable quality translations out the door quickly, easily and cheaply. Google has done well in designing a user-friendly platform that combines several CAT (computer aided translation) tools. The platform will undoubtedly gain popularity with professional and novice translators - good products that are free are hard to pass up.

This product does not have the broad appeal of some of its other products, such as Google search, Google Maps, Gmail and even the Google translate tool itself. These and other products are geared at the lay user and require no professional training to use. The Google Translator Toolkit on the other hand is an esoteric product which is geared towards professional users, people who are professional translators or those with specialized language skills. Moreover, the use of Translation Memories (TM) which is at the heart of the tool is the domain of professionals. People that do not own a CAT tool will typically not be able to access TMs in TMX format. The proof of the esoteric nature of the product can be attested to by the relatively quiet nature of the product’s introduction. The announcement appeared on some tech blogs and in news services that are associated with the localization and language industries, but went mostly unnoticed by the mainstream news services.

The impact that this tool will have on the translation industry remains to be seen. It can certainly be used for fast delivery of large projects and also for collaborative projects that require rapid deployment of translation teams working on the same project. However, workflows based on post-editing of MT have already been around for a while and are in use by many LSPs.

Some bloggers have pointed out that this tool will be used to crowdsource Wiki and similar data in order to make these articles available in more languages. But that is not a commercial application and would therefore have limited impact on the localization/translation industry.

Many LSPs may avoid uploading their TMs to Google Translators Toolkit. A peek at Google’s TOS (http://translate.google.com/toolkit/TOS.html) reveals that ‘by submitting your content through the Service, you grant Google the permission to use your content permanently to promote, improve or offer the Service.’ TMs are a valuable and proprietary resource and many companies may not want to pass it along to Google. And besides, many of the biggest companies have invested considerable funds to join the TDA (TAUS Data Association) initiative for TM sharing and would probably not consider jumping on the Google TM sharing wagon where they have zero control over the use of their data.

Google’s objective in this offering is to get their hand on as many TMs are possible. TMs are aligned sets of translated corpora, usually of very high quality, and would give them an advantage in training their SMT engines enroute to improving their MT quality.

Will the toolkit help Google achieve MT which is as good as human translation? For years computers could never beat humans in chess. But then in the 1990’s Deep Blue beat the world champion Kasparov and since then humans just don’t have a chance. If Google does to translation what computers did to chess, many companies and people will have to look for other kinds of work.

The following is a short description of the tool’s workflow:

To start a project, you need to upload the file you want to translate (selection of multiple files are not supported). The basic file formats or DOC, RTF, TXT and HTML are supported. You have the option of uploading a translation memory (TM) in TMX format, or selecting the global TM which is basically all the aligned translations that Google has stored on its server. You can also upload a glossary of terms. If you upload your own TM, you can keep it to yourself, to a select group of collaborators or share it with everyone.

Screen shot of Main Panel

Once you have uploaded the document, web page, Wiki article or Knol you select the language pair and then translate the document. Currently, only English source documents are supported so if you need to translate documents into English, the toolkit will not support it.

Upload Document for Translation screen shot

After Google translates your document, the source and translation are displayed in a side-by-side format. Translation units (Tus) are segmented as sentences. You can select a TU by clicking the mouse inside the left-hand pane on the TU you wish to focus on. The source TU is highlighted and the translation is displayed in the editor box which allows the translator to post-edit the translation. The translator can easily move from TU to TU using the Next/Previous links. This makes editing the document a snap.

When using the tool to post-edit translated web pages, the tool extracts all editable text that is not a part of the body text, such as pulldown menus and buttons, and displays this text at the bottom of the editor. This allows you to edit this text as well. However, the tool does not extract meta tag text and does not allow you to edit this within the tool editor.

Editing box screen shot

An optional toolkit box which can be displayed at the bottom the window displays TM matches if applicable. Exact matches (100%) are indicated, and fuzzy matches are displayed as well, although the degree of fuzziness is not indicated. Users can also access a concordance (Dictionary) function as well as view Glossary matches if any were found.

Optional Toolkit box screen shot

Once the editing has been done, the translated document can be saved and downloaded. All documents are converted initially into HTML format, which is the native format which the Google Translation Toolkit uses. The document is reconverted after saving it and is downloaded in its initial format.

Etilize is the worlds largest source of product information with aggregated data on over 7 million products from 20,000 manufacturers in 30 countries and in 20 languages. The company website, a href=http://www.etilize.comhttp://www.etilize.com/a, has been localized into 5 languages by GTS and is now available to readers in English, French, Italian, German, Spanish and Dutch.

Companies in all industries are using GTS for website localization and translation services. Localizing a website is a cost-effective way of reaching customers internationally.

GTS has set up a jobs feed on Twitter which you can see on a href=http://twitter.com/GTS_Jobshttp://twitter.com/GTS_Jobs/a. This feed is updated frequently with the latest job offers from GTS. GTS has a constant need for translators in many language pairs. To get our job updates, just login to Twitter and follow GTS_Jobs. You can set Twitter to send the job updates to your cellphone or by email.

If you dont have a Twitter account, getting one is easy from the Twitter website. Are you a freelance translator? Follow the GTS Jobs tweet to get the latest on work opportunities.

SDL, the leading supplier of translation tools and one of the biggest translation companies in the world announced a global strategic partnership with Statistical Machine Translation (SMT) developer Language Weaver. See http://tinyurl.com/c7vtrt for more details.

But what does that indicate about the current state of the translation industry? 2007 was an excellent year for translation companies, and 2008 was also a very good year. But 2009? No official financial results are in, but by all indications it is turning out to be a bleak year for the big players in the translation industry. With the US economy shrinking by 6.1% in Q1 2009 in what is reported to be the biggest slump in 50 years (see http://tinyurl.com/d4qt4f), translation companies can not be expected to escape unscathed by the global downturn. In fact, the translation industry should be hit even harder than other industries as companies are scaling their budgets on non-essentials.

Lionbridge, one of the biggest translation companies in the world, announced a 8% reduction of its workforce in March 2009 (see http://tinyurl.com/djqa2g). It stands to reason that most or all of the other large companies performed similar workforce reduction moves to meet the lower levels of demand and to avoid a cash drain which could put them in jeopardy.

Many freelance translators are also hurting. Over 80% of human translation work worldwide is done by freelance translators who generally work from home. It appears that work is in short supply right now and many freelance translators are spending much of their time soliciting new business (mostly by spamming translation agencies–GTS receives an average of 20 inquiries a day, up 300% when compared with 2008).

The global downturn is pushing the large translation companies to develop machine translation technology that can lower prices and reduce budgets for their clients. Enter SDL and Language Weaver, whose collaboration is a move in that direction. Another factor which is adding uncertainty to the Machine Translation industry segment is the presence of Google, perceived to be the best MT tool in the market. Google has not revealed any kind of business plan for their translation tool and as long as this service is offered for free, companies will face an uphill battle in trying to sell the same service for money.

HTH is leading hotel group consisting of the Pacific Beach Hotel and Pagoda Hotel on Oahu and the King Kamehameha’s Kona Beach Hotel on the Big Island. GTS translated the websites for the Pacific Beach and Pagoda hotels into Japanese, including translation of all web pages and Flash graphics.

You can see the translated websites on: www.pacificbeachhotel.co.jp and http://www.pagodahotel.jp.

Red Herring, the leading technology media group that covers startup companies, recently announced their 100 finalists for the coveted Red Herring Europe award. One of the finalists is Spanish company Ta with you, a startup from Barcelona that is developing SMT (Statistic Machine Translation) solutions.

The company website, http://www.tauyou.com, advertises two types of translation solutions:

* An SMS solution for translation of text between several languages. The user enters text as an SMS on his/her cellphone, specifies the language, sends it to a phone number in Spain and gets back the translation. Information on use of this solution is described on www.tauyou.com/en/sms.html. According to the Ta with you blog, this service is offered in partnership with Vodaphone.

* T-Image solution. The user takes a photo of some text using his/her cellphone, uploads it to the Ta with you website and gets back a translation of the text. This solution can be accessed on http://www.tauyou.com/m_new/image.php?idioma=en. Further information on this feature can be found on the Ta with you blog. A description of the T-Image API can be found on http://tauyou.3scale.net/.

Another solution which is touted on the Ta with you blog is real-time translation of TV show subtitles. This feature would work on a film that has subtitles in one language, making them available in other languages in real-time.

Disclaimer–the SMS and T-Image features did not work when tested by us. If anyone has experience with these services or can successfully test them, please let us know. Furthermore, Ta with you does not provide any access to their SMT software for purposes of evaluation or testing. The company provided a translation portal for the 2009 Mobile World Congress in Barcelona that combines machine translation and TTS playback (see it at http://gsma.mobitra.mobi/). However, this portal uses Google MT and does not use the Ta with you MT system.

The following is an interview with company CEO Dr. Diego Bartolome, which was conducted by e-mail on March 23, 2009. A bio of Dr. Bartolome can be found on http://younoodle.com/people/diego_bartolome.

GTS: Please provide more information about your proprietary SMT system. What
languages does it support? How accurate is it?

Diego Bartolome: Our proprietary SMT solution is a domain adapted statistical machine translation (SMT) platform, which is completed by pre- and post-processing steps to guarantee quality in particular verticals. Our technology has a human quality for a particular domain, thus significantly better than any other free system for any language pair.

Although we have a basic translation engine fed by freely available multilingual corpora from the Internet, as well as dictionaries and other resources, we basically work really close with our clients and make a system specifically designed to fulfill their needs. After an analysis of the domain and pairs of languages they require, we need at least 1 million words from their previously translated texts or translation memories for a particular client and domain e.g. medical, pharma, bank, news, computer, automotive, etc. The higher the amount of data for that domain, the better the overall quality of the system. This and additional data we generate feed the SMT engine, and we carry out the training for the domain to achieve almost human quality or that task.

Therefore, we work with any language and any domain, achieving a close-to-perfect accuracy for the domain.

GTS: Can the Enterprise server be deployed by the client? Do you sell this software? Do you provide remote service through SOAP interface or equivalent?

Diego Bartolome: Yes, domain-adapted SMT is mostly an enterprise-level solution. We build the system for the client as I have explained, and we can deploy either as a Software as a Service in an external server or within the client premises, which can be accessed through web interfaces or as any means desired by the customer, we are IT specialists.

However, we need access to the system to enhance the quality monthly.

The business model is a set-up fee plus monthly maintenance and support, and the price depends mostly on the client.

GTS: What is the throughput of your MT solution?

Diego Bartolome: For the domain adapted SMT, we can translate more than 40 million words per day with a single server, with full scalability. We can fulfill the language needs of any company, reduce their costs by more than 40% and decrease delivery times to the minimum.

GTS: Is there any way to test the Enterprise software? Can you provide a web page
for testing purposes to test text translation?

Diego Bartolome: Solutions for the customers are protected by NDAs, so we cannot offer them to other people. However, for sure you can test a Catalan to Spanish translator if it helps, just send an email to translator@mobitra.mobi, with Catalan to Spanish or Spanish to Catalan in the subject, and the text in the body. To see how it works, you can look at the Catalan - Spanish bilingual newspaper www.elperiodico.es / www.elperiodico.cat

GTS: On your website, you advertise a ‘T-Image solution’ which allows mobile phone users to take a photograph of text and upload it your website to get a translation. How does your server read the image, does it have an image to text engine? Does it have error correction software and how clear does the text have to be, how large does the text have to be?

Diego Bartolome: For our T-Image! product, it currently works fine if the customer 1) avoids text distortion and perspective, 2) holds still and uses good lighting, 3) turns off the flash, 4) fills the screen with the document (best if portrait or landscape), 5) takes a sharp and focused photo. Besides, it’s recommended that the camera has autofocus capabilities if they want to take a picture of small texts (e.g. 12 or 14 points), otherwise the font size shall be very big (greater than 30 points). At this point, we do not support handwritten texts, although it might work depending on the way the clients write.

Our technology here is mostly related to the quality increase of the image and the robust recognition of the text in a natural scene. Since the product is in beta phase, we support the recognition and / or translation of printed materials such as newspapers, books, cards, menus, etc.

For the T-Image!, the average response time including the upload of the image is about 20 seconds.

GTS: How does the automatic subtitling solution work? Do you need at least one subtitle to exist? Does it do it in real-time? or do you need a pre-processing phase.

Diego Bartolome: The real-time subtitling requires at least one subtitle to exist, and it is done in real time. Therefore, if the TV channels deploy it, it will be something they preprocess and then provide the viewers with a set of subtitling options, and if it’s deployed at home, the subtitling will be always done in real-time. Since it’s an ongoing project, we prefer not to disclose many details …

GTS: Thanks very much for this interview.

A seemingly new player has been recently detected on the Machine Translation scene called Lucy Software (http://www.lucysoftware.com/). Their Kwik translator, which you can test online on their website, provides automatic translation between numerous language pairs. Two languages which they offer, Catalan and Galician, are not handled by any of the other MT vendors (except Google). But this is by no means a new product, but rather a very old one. In fact, this company is like a hot potato which has been rebranded and sold every few years. Its exact history is unclear, but this technology and group is of German origin and started as a project at Siemens. At some point, there was a management buyout (MBO). In recent years the company was called Comprendium. In 2006 it was acquired by Braintribe (www.braintribe.com). It was recently rebranded again to Lucy Software.

It is unknown why this frequent rebranding has occurred, but the company is built around very solid rule-based MT technology (RbMT) with highly advanced lexicography tools. In fact, they have a software add-on called Lexishop which is probably one of the most powerful customization tools available for any MT system around (an excellent article on their technology can be found on http://www.mt-archive.info/Comprendium-2004.pdf). Their main customer is German dictionary publisher Langenscheidt which sells a line of German translation software tools based on Lucy technology.

The importance of getting a translation 100% correct can never be over-stated. Even the slightest deviance from the original text can convey an entirely different message to the one that was intended; and this only serves to underline the value of sourcing good quality, professional translators.

But what about translations that are 100% wrong? Well, besides the embarrassment suffered by the parties involved and the bemusement/amusement of those looking on, the main thing it reveals is that they have simply cut corners and haven’t bothered to do their foreign language homework.

And there are many horror stories out there which act as a gentle reminder of the consequences of making a translation blunder.

In 2008 alone, there have been many high-profile tales of translation woes. There was the restaurant in China that, whilst obviously trying to make its shop-front more appealing to the English-speaking world for the Summer Olympics, decided to use an online translation tool: ‘Translate Server Error’ was the resulting message, designed to ‘entice’ anglophiles through its doors.

Then, there was the road sign in Wales that, when translated from Welsh into English, read: “I am not in the office at the moment. Send any work to be translated”.

And now, as 2008 draws to a close, German Research Institute Max Planck has just published its latest Forschung journal, with China as its main focus. With the best of intentions, its front page was adorned with what they thought was a classical Chinese poem; it turns out it was actually an advert for a brothel.

So, to use a well-coined expression, it really is best not to get ‘Lost in Translation’ when it comes to important business messages. But when it does happen, nothing could be quite so funny.

© Lingo24 Translation Company. This entry is a contribution by Paul Sawers of Lingo24, a leading translation services company. Check them out on http://www.lingo24.com.

GTS has announced the release of a new website translation system which streamlines the website translation and maintenance processes. By launching an advanced system which is the first of its kind on the market, GTS takes the headache out of the website translation process, making it easier than ever for website owners to get their messages out to an international, linguistically diverse audience. And due to our unique workflow, customers get top quality website translation at affordable prices and at huge costs savings.

System features:

- High quality human translation

- Languages: French, German, Spanish, Italian, Chinese, Japanese, Arabic

- Cached web pages that appear in foreign language search engines

- Automatic version tracking-changes in English language website are updated in all languages

For more information and a price quote, please send an email to info@gts-translation.com.

GTS has received an order to translate a website for Hypo Fondsleitung (Liechtenstein) AG, which is a subsidiary of Hypo Investment Bank (Liechtenstein) AG. The website address is http://www.hypofunds.li/.

GTS provides a wide range of translation services for banks and financial institutions, including translation of website content, financial and legal documents, purchase agreements, statements and more. In 2006, we translated from German to English a book entitled “Suspicion of Money Laundering - In the Crossfire of International Due Diligence Obligations” for the Hypo Investment Bank.

GTS partnered with Flowserve (NYSE: FLS) to provide their Careers website into eight languages: French, German, Italian, Spanish, Dutch, Portuguese, Simplified Chinese and Japanese. The Flowserve Careers website can be found on http://www.flowserve.com/careers.

The project posed several challenges, since many languages need to be delivered in a short amount of time. Various types of files were translated, including Flash, HTML and PDF.

testing

HTH is leading hotel group consisting of the Pacific Beach Hotel and Pagoda Hotel on Oahu and the King Kamehameha’s Kona Beach Hotel on the Big Island. GTS translated the websites for the Pacific Beach and Pagoda hotels into Japanese, including translation of all web pages and Flash graphics.

You can see the translated websites on: www.pacificbeachhotel.co.jp and http://www.pagodahotel.jp.