Guest post: 10 things translators need to know about machine translation 2

Meet Gwenydd Jonesa freelance Spanish to English translator and professional trainer. She has two MAs, the first in Translation Studies and the second in Legal Translation, and the DipTrans (CIOL). With 10 years’ experience, Gwenydd specializes in business, marketing and legal translation. She is also a copywriter.

Learn more about Gwenydd and some of the courses she offers by checking out her blog, translatorstudio.co.uk.


I don’t know about you, but I spend much of my life going from one translation project to the next. I want to learn about translation technology, but am always putting it off. Not my idea of fun. For me, m1074712_r55e018418b6a3achine translation is like the hologrammatic elephant in my home office.

Last June, I had to prepare a talk for the ProZ.com conference in Stockholm. Finally, an opportunity to confront the elephant. I set out to find answers to my questions, hoping to put my worries to rest. I wanted to find out: what is going on with machine translation? Is it a real threat to human translators? And if it is, what should we be doing about it?

In my webinar Your Essential Machine Translation Briefing, on 8 Feb, I’ll share what I found out. From the perspective of a technically challenged freelancer. In the first half, I’ll give you a summary of what is currently going on in machine translation. Then, I’ll share the strategies I’m employing in my work, to make sure I develop alongside automated translation. See you there!

In the meantime, here are a few things freelance translators need to know about machine translation.

  1. ‘Machine translation’ isn’t the same as ‘translation memory’ or ‘CAT tool’

Sometimes translators get these terms muddled up, which is understandable. A computer-assisted translation (CAT) tool is an application where you can write your translations. It uses translation memories (TMs) to keep a record of all your past work. Don’t confuse that with a machine translation (MT) engine. An MT engine is an application that automatically translates a text. You can integrate MT with your CAT tool, but they are two different pieces of software. They have different functions.

  1. Machine translation is more effective with certain text types

Since computers rely on data and rules, the more predictable a text, the better the output will likely be. Formulaic and simple texts work well. Creative and complex texts don’t. Life sciences, finance, IT and other technical genres lend themselves to machine translation. But while financial accounts are formulaic, an accountant’s blog is far less predictable. With the second, you may well be faster on your own, particularly if you use voice recognition.

  1. Machine translation engines get better results when they’re customised

Translators and companies that are serious about machine translation aren’t using Google Translate. They get their own machine translation engines and train them for a specific domain. They do this by inputting their translation memories. After that, they input more data on an ongoing basis, so the machine keeps adapting to them. This is how they get more accurate output. Then they post edit it and feed the final translation back into the machine. With a suitable text type, this helps them finish the translation faster than if they did it from scratch.

  1. Neural machine translation is a major change in the translation industry

The world of machine translation is starting to harness deep learning. This is based on neural networks. Neural networks have lots of uses in artificial intelligence. Language processing is one of them. So, computer scientists can use them to improve machine translation. Companies that are using neural machine translation include: Google, Microsoft and Facebook.

  1. Google Translate is now using neural machine translation in some language combinations

Google recently announced that it is using neural machine translation in Google Translate. For now, it is limited to certain language combinations. They rolled it out with a total of eight language pairs. All are to and from English, combined with French, German, Spanish, Portuguese, Chinese, Japanese, Korean and Turkish. It’s free, but remember the data is public, so you can’t use it if you’ve signed a confidentiality agreement. Google Translate isn’t customised (see point 3), but it has lots of data. The jury is still out on how good the updated tool is.

  1. You can subscribe to your own machine translation engine and train it

You may not realise that companies that develop machine translation engines sell subscriptions. You can even pay to train your own engine using your translation memories. Post-editing isn’t just about an agency sending you texts. You can learn how to post edit, get a customised engine and then do whatever you want with it. The profits and control will be all yours.

  1. Companies that sell machine translation are battling for your business

If you want to try machine translation, you have to go shopping for a provider. The different companies that offer machine translation solutions (including SDL, Lilt and Systran) publish data to show how effective their software is. It’s all quite technical and confusing. And it can be biased. You can go to TAUS and the eMpTy Pages blog, for unbiased information.

Perhaps, like me, you prefer to see for yourself. One way of doing this is to observe yourself for a month in your work, and see how many words you average per hour. Then, pick whichever machine translation software takes your fancy and use it for a month. Track your turnover to see whether your hourly average gets faster or not. Then you’ll have your own data to tell you whether it’s worth continuing to invest.

  1. Machine translation isn’t currently replacing human translators

Improvements in machine translation don’t mean we’re all out of a job. But, it may mean our jobs will start to change. Commercial translation is often about getting an acceptable translation as quickly as possible. You can complete some texts faster by using customised machine translation and post-editing. And sometimes that is what the customer wants. As machine translation continues to get better, we can expect demand for post-editing to grow.

  1. Machine translation pricing isn’t set in stone, yet

While machine translation has a long history, it’s still being consolidated in our industry. Lots of translators and translation agencies are struggling to get their heads around it. Some translators are concerned that post-editing means lower pay. Certainly, ruthless agencies will try to use it as another way of driving prices down.

But, that doesn’t mean smart translators can’t use it to increase their profitability. Why shouldn’t we earn more on the days we work as post editors? If we learn about it, and talk about it, we’ll soon know how we want to price it. We’ll know when to refuse a job. Translators can choose to accept post-editing jobs only when they’re going to make more money for their time.

  1. Freelance translators have options regarding machine translation

We don’t all have to go running for the hills before the robots attack. Becoming a post editor is just one option open to us. If you like the idea of it, you could post edit for agencies. But, you could also subscribe to your own engine and use it in your work. If you don’t want to post edit, there are a host of specialisation and diversification options.

Whatever path you end up choosing, now is a good time to get informed and come up with a plan. If major changes do take place in our industry, I for one will be ready for them. Ready to adapt. To continue being the one who controls my career. To protect my rates, serve the direct customer competitively, and understand the jobs (and prices) agencies offer me.

I encourage you to join me at my webinar on 8 Feb, Your Essential Machine Translation Briefing. I’ll share everything I’ve found out about machine translation, in simple, unbiased terms. I’ll also tell you the strategies I’ve come up with, and am now employing in my work. Sign up here!

This September, all roads lead to Curitiba! Reply

Today’s guest post author is Sheila Gomes – a freelance translator with over 20 years of experience who currently specializes in software localization and video games. Sheila is the manager of Multitude – an online information portal for translators and interpreters, and is one of the founding members and organizers of  TICWB – a networking group for local industry professionals.

Along with fellow freelance language professional and industry contributor Isabel Vidigal, Sheila is the co-organizer of this year’s ProZ.com regional conference in Brazil, which will take place this September from the 23rd to the 25th in the city of Curitiba. She shares her post today in Portuguese.


Minha primeira conferência de tradutores e intérpretes foi no Rio de Janeiro, em novembro de 2011: a III Conferência Brasileira de Tradutores do ProZ.com. Como foi a edição com o maior número de participantes até então, imagine o assombro da pessoa perdida entre mais de 300 colegas, com dezenas de apresentações e outras atividades para participar. Acabou virando a primeira de uma série: o bichinho dos eventos T&I tinha me mordido e hoje vou a todos que posso. Até chegar ao ponto de organizar em conjunto com a Isabel Vidigal o nosso evento do ProZ.com. A Isabel é veterana de eventos, já organizou inclusive a primeira Conferência do ProZ.com no Brasil, junto com a Rosana Malerba, em agosto de 2009. E agora o evento vem pra Curitiba, num dos poucos casos de saída do eixo Rio-São Paulo. Nesta minha cidade do coração, que acabou virando um polo de referência para tradutores e intérpretes por causa do trabalho ativo que temos aqui com iniciantes e veteranos, em vários projetos e ações. Estamos ansiosos e com vários planos para receber os colegas!

Assim como é para muita gente, o ProZ.com foi meu primeiro passo para conseguir clientes internacionais e fez uma grande diferença na minha carreira. Claro, é um grande recurso, mas funciona de verdade quando fazemos nossa parte, depois de estarmos preparados, de ter pesquisado o mercado e aprimorado as qualificações profissionais. O próprio site oferece uma série de ferramentas para isso, e tentei aproveitá-lo o máximo possível para aprender e também contribuir. Assim também é com a VI Conferência Brasileira do ProZ.com, que estamos organizando aqui em Curitiba entre os dias 23 e 25 de setembro: tentamos devolver um pouco do que conseguimos por meio do portal e oferecer outras oportunidades de fazer networking, receber treinamento, estabelecer discussões e momentos de socialização, para tradutores e intérpretes, iniciantes ou veteranos, e outros interessados na área.

Creio que uma das ações mais eficazes para mudar o mercado é dar acesso a iniciativas educacionais aos profissionais em formação e outras pessoas interessadas em ingressar nessa nossa área tão rica, mas também ainda pouco conhecida do grande público. É por isso que o desenvolvimento profissional inspira o tema do evento, “Boas práticas e caminhos”. Além de palestras e mesas-redondas, o evento oferecerá atendimento especializado individual ou em pares, na forma de miniconsultorias, para profissionais já atuantes e estudantes que buscam informações para se profissionalizar. E como a descontração é importante para estimular a integração dos pares, além do próprio evento, teremos encontros informais e passeios culturais.

Aliás, Curitiba é ideal para encontros assim, especialmente para tradutores e intérpretes, pois o que mais temos por aqui é: café! Espaços simpáticos, pitorescos, convidativos a cada esquina, dos maiores e festivos aos menores e aconchegantes, não faltam lugares para todos os tipos de grupos ou apenas para um bom papo entre duas ou três pessoas. E para quem vem, mas já sabe que pode ter que trabalhar também, praticamente todos os espaços oferecem wifi, além de alguns outros espaços de acesso gratuito como a biblioteca pública (a uma quadra do local do evento) ou algumas praças. Isso sem contar restaurantes, bares, espaços culturais e outros eventos para conhecer e investir no networking até fora do evento.

É por essas e muitas outras que esperamos você aqui: em setembro, todos os caminhos levam a Curitiba!


Meet Sheila and all of the excellent speakers who will be present at this conference – like keynote speakers Marta Stelmaszak and Paula Ribeiro – by registering today on the main event page: http://www.proz.com/conference/686

Registration fees can now be paid in the local currency! The early bird price has been extended so those who are interested in paying in reais at this discounted price may do so. Don’t delay! Prices increase in just a few short days, on July 23rd. More information about paying locally can be found on the event page under the “Opção de pagamento em reais” heading.  

Want to learn more about what to expect at this conference? Program highlights are featured in this short video:

From the corporate corner: Let’s tell our story 4

Meet Lori Thicke: founder of Lexcelera and the non-profit organization Translators Without Borders. In this guest post from the corporate corner, Lori speaks on why translation is under-appreciated and what we can do about it.


New York City at the height of the Ebola crisis in West Africa. I am speaking about language to a roomful of high-level executives from the largest aid groups, convened as part of a series of UN focus meetings.

I cover communications in the Ebola crisis, and how utterly unhelpful it is to tell people how to avoid Ebola in a language they don’t understand. After all, you wouldn’t go to France with public health posters in English: why would you do so in Liberia?

Afterwards, the Executive Director of one of the world’s top aid organizations (you’d know the name) says to me, “We really hadn’t thought about that.”

Hello, what? You didn’t think that it was important to talk to a rural villager in her own language? That language wouldn’t matter much, even when you’re trying to stop an epidemic as perilous to the world as Ebola?

Here’s a news flash: communicating in the wrong language is not communicating at all.

Lori Thicke: CEO and founder of Lexcelera

Humanitarian groups not getting that simple fact is the main reason I founded the translation charity Translators without Borders. Yet the same ignorance about how important language is also bedevils anyone who earns a living in the translation industry.

Before Translators without Borders, I founded a language company, which I operate to this day. Lexcelera began life in Paris, France, and we have a few small offices now on three continents. But operating in a different world, in business, in communicating B2B and B2C, we still face the same issues as in the humanitarian sphere: translation is wildly, crazily undervalued.

It may seem strange to make the leap from humanitarian translations to the business world, but I believe the same core problem affects both: people outside our industry, whether nonprofits or companies, think they can get by just some token translation. I mean, have you ever seen how most companies do their international customer support sites? You might see the menu items in a few main languages, but the information itself is in English.

The assumption there, of course, is that everyone speaks English. Talk about wishful thinking!

In the commercial sense, this wishful thinking translates into undervaluing our services – and that in turn leads to commodity (read low) pricing. This commoditization springs from the idea that what we do isn’t worth very much, so any old provider will do as long as the cost is cheap enough.

I can’t think of another industry where prices go down, year after year.

This may be a contrarian view, but I see the huge investments that are being made to improve machine translation (MT) as the one acknowledgement that speaking to people in their own language is the only way to go, and that technology is needed because there are too many languages and too much content.

Wait, investing millions and maybe billions in machine translation is actually recognition of the value of our work? Yes, that’s what I believe. But as I said, that is no doubt a contrarian view.

In any case, MT is really an aside to the bigger issue: the lack of recognition of the value that professionals bring to multilingual communication.

I believe the only way we can fix this is by telling a better story. A compelling story. Somewhere along the line we stopped being visible. When was the last time you saw a translator in a movie? In the press? We are one of the professions you don’t see or hear a lot about. And that hurts us.

We need to take control of the narrative.

ProZ.com and other professional bodies could help here by relentlessly passing the message that in our increasingly borderless world, companies need our services in order to communicate better – and to sell better.

Our trade associations could make headlines with stories about how people are more likely to buy products and services when addressed in their own language and how companies grow more when they get language right.

These stories could be backed up by hard numbers, compelling statistics that tell the story of happy customers and engaged employees. For example, the Common Sense Advisory tells us that people are 6 times more likely (duh) to buy from a website when addressed in their own language.

Citing facts like this can make the case that translation is not a commodity but an investment where quality pays.

I believe we need to tell our story as publicly as we can to raise awareness and appreciation for our craft. Translators need to be linguists, they need to be subject matter experts and they need, almost above all, to be good writers. This is a unique and valuable skillset that allows professionals to craft a translation that does the job it’s supposed to do: communicate a message that will be understood.

Now, is that so hard to understand?

 

Guest post by Paula Ribeiro: Interpreting the present to translate the future Reply

Today’s guest post was written by Paula Ribeiro – president and co-founder of the Portuguese Association of Translators and Interpreters (APTRAD). This organization will be holding its first international conference on June 18th and 19th in Porto, Portugal.


APTRAD, the Portuguese Association of Translators and Interpreters, was established in February 2015 by a group of freelance professionals in response to a perceived need for a modern, creative and innovative approach in order to achieve greater cohesion and exchange of information at a national level within the profession. After almost one year of hard work we are proud of achieving some of the important goals we initially set.

APTRAD’s motto – Interpreting the present to translate the future – reflects the Association’s aim to promote and foster the growth of its professional members, and to support the integration as professionals of all future translators and interpreters into the market.

Pursuing this thought, APTRAD is holding its first International Conference on June 18-19, 2016. Taking advantage of the main festivities of our city in that month, we will try our best to turn this event into a big party, welcoming all translators, interpreters and linguists in general to join us in our beloved hometown – Porto.

The theme of the conference will be “Stages in the career of a freelancer” and will tackle the different phases in the career of a professional freelance translator and/or interpreter and what’s expected and required at each stage. We will have renowned speakers who will certainly inspire all of us with their knowledge and experience in several areas of our profession.

The organisation of this event becomes much easier with the valuable help of our partners in which ProZ.com is included as an essential reference in the career of so many professionals. A big thank you on behalf of APTRAD.

Feel free to visit our website at www.aptrad.pt and more specifically the conference website at www.aptrad.pt/conference/conference and drop us a line if you need help from us. See you in Portugal, in June!


About Paula RibeiroPaul Ribeiro

Paula Ribeiro started translating in 1997, and since then she knew that this was the career she wanted to pursue! She graduated in 2006 with a Master’s Degree in Specialized Translation and Interpretation with English and French as her working languages, and later Spanish as her third language. She is currently pursuing a postgraduate degree in Computer Assisted Translation.

In 2010, Paula decided to create her own company – Crossingwords – and to undertake translation and interpretation as her main occupation, always maintaining her education and training as a key part of her professional and personal development.

As an event organizer, Paula has planned several conferences on both a national and international scale, including the 2013 ProZ.com International Conference.

Since February 2015 Paula has been one of the founders and the President of APTRAD, the Portuguese Association of Translators and Interpreters, a formally constituted non-profit organisation based in Porto.


Did you know?

You can find APTRAD’s international conference listed on ProZ.com’s translation industry events calendar, along with dozens of other language and translation-related events that are scheduled to take place this year, ranging from workshops or seminars, to powwows, to regional events, to major international conferences.

See the announcement: http://www.proz.com/topic/298930

Guest post: Why I volunteer for Translators Without Borders Reply

Pieter_Beens
Pieter Beens is a freelance translator and copywriter working in English to Dutch, and a frequent guest contributor to the Translator T.O. 

In this post, Pieter shares his experience as a volunteer translator for Translators Without Borders.


I just completed a translation for Translators Without Borders, my fourth this year. And I must admit I was touched. This time I translated for a charity that helped orphaned children get back to school after the Ebola outbreak last year. Such a beautiful initiative needs our support. I did my small part by translating their sponsoring letter into Dutch, and hope that the letter will help raise the funds necessary to bring these children back to education. That is why I chose to register as a volunteer for Translators Without Borders a couple of years ago, and why I have already translated more than ten thousand words through this organization for several different charities. And there are many more volunteer translators doing the same, donating their time and effort towards helping various other charity initiatives that deserve support. Through Translators Without Borders, we have already translated 30 million words for a multitude of audiences in almost every country in the world.

About Translators Without Borders

Many of us know Doctors Without Borders, an international organization offering worldwide medical support in the event of humanitarian crises and other urgent situations. In 1993, two pioneers in the translation industry founded a linguistic equivalent of it, Translators Without Borders, aimed to link translators around the world to vetted NGOs that focus on health, nutrition and education. Today the platform is affiliated to ProZ.com and sponsored by many translation agencies worldwide. Translators Without Borders offers them a chance to share their knowledge and resources in order to help the needy, while at the same time sponsoring can show off their social responsibility. The translation agencies do not necessarily offer translations, but they offer funding. Translations are done by professionals who voluntarily sign up to offer their help to organizations in need of translations in their language TWBpairs.

Registering to volunteer your services through Translators Without Borders does not mean you are obligated to accept every project that comes your way through this organization, nor does it necessarily guarantee that projects will be passed to you. As you can imagine, the demand for volunteers varies greatly depending on language pair and pool of available candidates. Indeed, there is a very high demand for professionals working in certain pairs, and less demand in other pairs. There may also be many translators volunteering in some language combinations, and far fewer volunteers available in others.

Why choose Translators Without Borders

Last year I wrote about five reasons to translate for charities and tips for supporting charities as a translator. Translating for Translators Without Borders can be seen as a part of my commitment to offer my professional services to organizations that support those in need. At the same time, Translators Without Borders does not require a huge commitment. In my language pair (English into Dutch) requests are sent irregularly, from organizations like Wikipedia, street newspapers, and the International Red Cross. The nature of translation tasks varies from interviews, to fundraising letters and other important information about diseases like the Zika virus, for which I recently translated a text.

In general, project deadlines can be fairly long; in many cases the deadline for a text with 500 words may be around 10 days, while the deadline for texts with 2000 words can even be 30 days. That enables translators to focus on their important tasks and to do volunteer tasks in their own pace. After having delivered the text many clients often leave gracious feedback, knowing that without our help it would have been much more difficult to reach local audiences in their local languages.

In short, volunteering for Translators Without Borders is a rewarding opportunity that enables freelance translators to use their professionalism and passion for a higher goal. I highly recommend it!


Did you know?

CPN_logo
Members of ProZ.com’s Certified PRO Network do not need to undergo any additional screening process to join Translators Without Borders’s team of volunteers.

You can learn more about this initiative and apply for inclusion in the program here: http://www.proz.com/pro-tag/info/about/

Post-editing: Blessing or curse for translators? 1

This is the second post in a two-part guest blog series by ProZ.com professional trainer and conference speaker Federico Gaspari. The first post in this series can be found here: “Machine translation: Cause or solution of all evils?”


One is unlikely to make many friends among translators talking about machine translation (MT) – unless the conversation is restricted to deriding its stupid mistakes and emphasizing its uselessness. A related topic that is possibly even less popular than MT among translators is post-editing (PE), also because it’s less easy to come up with funny stories of hopeless mistakes. Let’s face it: while pretty much everybody with at least a modest knowledge of two languages can be amused by the sarcastic appreciation of what is lost in (machine) translation, deriving pleasure from blunders occurring when post-editing MT output is a rather more subtle activity, whose enjoyment requires much more effort. This post discusses some issues concerning MT, translation quality and PE, focusing on some current trends in the translation industry of interest to professional translators.

Translators, MT and PE

Surprising though it may seem, there are dozens of threads on MT in ProZ.com’s technical forums, and one finds a mixture of (mildly) positive and (extremely) negative opinions, depending on the experiences of the community members who have posted their views. One of these forum threads, entitled “What’s your opinion on machine translation and quality?” has attracted one of the largest numbers of replies (more than 130) and views (over 16,000) of all the threads in ProZ.com’s technical forums. This incredibly popular thread is particularly close to my heart, because Daniela Zambrini initiated the discussion to announce an invited talk on MT and PE that I was due to give a few weeks later at the  ProZ.com 2014 International Conference which she organised in Pisa, Italy.

I’m under no illusion that I was responsible for the amazing popularity of the thread: in fact, Daniela’s well-intentioned post attracted replies which mostly ranged from outraged to exasperated, so much so that I was having second thoughts about whether I should actually go to the conference and give my presentation on MT and PE. Making many new translator friends had not been a consideration in accepting to give a talk at the conference a few months before (I already have quite a few of them, and we normally avoid discussing MT and PE…); but as the event was getting closer, I didn’t fancy the prospect of facing a particularly hostile and aggressive audience of angry professionals. As it turned out, my 45-minute talk at the conference in Pisa was rather well-received (in fairness, I smoothed over some of the contentious points that were likely to get on my listeners’ nerves…), and it was followed by a very civilised and interesting Q&A session at the end.

I even enjoyed some one-to-one conversations with translators who had listened to my talk and approached me during the rest of the conference: on the whole, they were genuinely curious about MT and PE, and I appreciated their honest questions and comments on these inevitably sensitive topics. In addition to a general curiosity to understand how MT works, several delegates at the well-attended ProZ.com 2014 International Conference in Pisa showed a keen interest in learning more about PE. As part of these conversations, some translators reported that they had been approached by LSPs and agencies as well as by direct end clients with requests for quotes for PE. As a result, these professionals were considering whether they should start offering PE services in addition to “standard” translation jobs, but they had no idea of the skills required and of the rates that they should charge. This blog post gives me the opportunity to discuss some issues related to PE that can be of interest to a wider audience of professional translators who are at least open to the prospect of securing PE jobs.

Post-editing MT output is different from translating and revising

At the risk of stating the obvious, it should be made clear that PE is very different from translating and revising translations done by (junior) human translators. The main reason for this is that MT systems make mistakes that are very different from those made by professionals, including relatively inexperienced ones. In addition, MT systems come in many shapes and forms: alongside the traditional rule-based approaches, statistical architectures are now particularly popular; these two basic types can be combined to obtain hybrid systems, and some researchers are now experimenting with neural MT, a new paradigm that seems to hold great potential for substantial improvements in output quality. Each of these types of MT systems is more likely to make certain kinds of mistakes rather than others, calling for different PE interventions.

In addition, different resources are required to develop MT systems with these approaches, and their output varies dramatically depending on the amount and quality of the available resources. A related crucial variable is the language pair involved: in principle, some approaches to MT system design are more promising for certain language pairs than others. However, the technological expertise and resources available for MT system development are unevenly distributed: while abundant human and technical resources can be tapped into for some languages (e.g. English and other widely used European languages as well as, increasingly, a few additional major world languages such as Chinese and Arabic), most languages are not well served at all by MT due to the lack of appropriate resources. There are techniques to deal with these shortcomings, but they are not always very effective.

One case in point are the huge sentence-aligned parallel corpora required for the development of statistical MT systems, whether they belong to the phrase-based or to the syntax-based category; while LSPs and freelance translators possess vast translation memory databases containing high-quality translated texts for certain language pairs, the data sets available for many others are far too small to offer the critical mass needed to kick-start the development of effective statistical MT systems. In practice, this means that the quality offered by MT systems (whatever their design) for several language pairs cannot yet be acceptable. This in turn determines whether PE is a reasonable proposition for the language pair under consideration or not. A closely related variable has to do with the text type in question: for some particularly challenging text types (even within the technical and specialised fields, say medical reports and legally-binding rental contracts) it may still be impossible to develop decent MT systems, e.g. due to the lack of relevant training data such as in-domain sentence-aligned parallel corpora in digital format, which can be very difficult to come by for certain language pairs in highly specialised and sensitive technical domains.

Many forms of post-editing

One common, but erroneous, assumption is that there exists only one type of PE; however, this is far from the truth. In fact, various PE levels can be appropriate for different purposes, given specific circumstances: at one extreme, light or minimum PE involves fixing only major errors, e.g. those that make the MT output incomprehensible or misleading (vis-à-vis the input in the source language), whereas stylistic nuances or relatively minor imperfections can be tolerated and do not require any correction – in other words, one is prepared to accept a less-than-perfect final target text, which can be good enough, for instance, for ‘gisting’ or information-gathering purposes; at the opposite extreme, there is complete or maximum PE: in this scenario, on the other hand, every inaccuracy in the raw MT output must be corrected, polishing up all minor details, i.e. the aim of complete PE is to obtain a final target text whose quality is equivalent to that of a professionally translated text. Note that, while professional translation invariably aims at delivering top-quality target texts, (light/minimum) PE can be carried out with the much more modest ambition of providing a final text that is usable in certain circumstances, accepting that it may be (very) far from perfect.

While this division may sound intuitive in theory, applying it in practice is quite complex. First of all, there are many intermediate cases between these two extremes of light/minimum and complete/maximum PE, and one has to determine which level of PE is most appropriate to a specific scenario, depending on the needs and expectations of the translation’s end users. This is a function not only of the time available for the PE job, but also of the initial quality of the raw output that is offered by the available MT system: even obtaining a final post-edited target text of average quality may require extensive PE interventions, if the initial raw MT output is particularly poor – in the end, the effort involved may not be worthwhile, compared to translating everything from scratch. Conversely, there may be cases where the raw output of a particularly effective MT system for a specific language pair in a well-defined textual domain requires only minor PE interventions to be brought to excellent final quality.

In short, the language pair and the text type in question, the design and quality of the MT system, the characteristics of the raw MT output and the intended use(r)s of the final revised target text interact in complex ways to dictate the actual level and effort of PE that are required. But this equation still leaves room for uncertainty from the post-editor’s perspective, as it is quite common for machine-translated texts to display uneven quality: for example, in a 10,000-word translation project, 10% of the raw MT output may be (nearly) perfect with little or no need for improvement, 30% may be impossible to salvage even with extensive PE (i.e. one would be better off re-translating those entire passages from scratch), and the remaining 60% may require different forms of intermediate PE (say, within the same paragraph one preposition must be changed in a sentence, a final ending agreement in another, but a whole dependent clause turns out to be wrongly translated and completely incomprehensible elsewhere). It is easy to see that PE can become a demanding activity, and the effort it requires in terms of skills and time is often difficult to predict and convert into clear rates that can be charged to clients with a transparent pricing scheme.

Factors to be considered when offering post-editing services

Still, with the increasing adoption of MT in professional translation workflows, the demand for PE is rising, so much so that many translators are considering whether they should offer PE services in addition to standard translation jobs. This is more likely, at least in the short term, for in-house translators of large LSPs that have the resources and expertise to develop their own customised MT systems for domains with constant demand from major clients, thus requiring some of their staff to take on PE roles in dedicated projects incorporating MT. But interestingly, some companies specialising in translation technology offer cloud-based “do-it-yourself” or self-service MT solutions that are accessible to freelance translators who are willing to invest in this area: this approach does not require extensive technical skills, because the training and set-up of the MT systems are guided in a step-by-step fashion for users with fee-paying accounts and managed at the back-end by the companies themselves. There are anecdotes of naïve clients looking for easy discounts who generated garbled output with free online MT systems, asking translators to fix the inevitable errors at cheap rates; however, since free web-based MT services are not customised to specific domains, but they are one-size-fits-all systems, this approach is unlikely to be successful: it is rather pointless, if not counter-productive, to carry out PE if the initial quality of the raw MT output is very poor.

Hence, even before considering the possibility of offering professional PE, one must be sure to have at least a decent-quality MT system available. Although it is very difficult to generalise, all else being equal (e.g. the domain and level of technicality of the source text, the quantity of language resources available for system training and development, etc.), MT into English (from, say, German, Russian or Chinese) tends to give better results than the opposite translation directions, i.e. from English into these target languages. As a result, in principle technical and specialised translation projects into English should be good candidates to explore the potential benefits of combining MT and PE. Although techniques for MT quality estimation are improving, it is still very difficult to accurately predict in advance the quality of raw MT output that will be obtained for a specific source text, and especially if this will be viable for subsequent PE. One must try and see whether PE (at the level required to obtain the expected final quality) is faster and more efficient than translating from scratch, e.g. with translation memories in a standard CAT environment. If they are open to this possibility, translators are well placed (more so than their clients) to gauge whether incorporating MT followed by PE in the translation workflow for specific projects can result in time gains and, potentially, in more competitive rates.

Open issues with PE

Some LSPs and freelance translators (including ProZ.com members!) have started to offer PE services, admittedly of the complete/maximum type, where the explicit goal is to deliver a final revised target text of excellent quality. Their pricing schemes vary depending on the language pairs and technical domains involved, and one open issue is whether PE should be charged pro-rata based on the regular translation fee, or by the hour: a quick survey of the online profiles of professionals offering PE services and of relevant discussion forums on ProZ.com shows huge variation in this regard, and there does not seem to be an industry-wide agreed approach yet. One crucial attraction of PE is that, given substantial volumes of MT-friendly technical material, one can in principle speed up turnaround times without sacrificing quality. With CAT tools and translation memory software increasingly integrating optional MT engines to process null matches, the practice of PE as part of technical translation projects is spreading quickly, and it may not always be easy to distinguish it from the editing of low fuzzy matches retrieved from translation memory databases: this in itself suggests that an honest discussion of the potential benefits of PE is timely and may prove in the interest of professional translators, so that they can offer clear and fair rates for their services, without relinquishing their negotiating power to budget-oriented clients.

Translators of today, post-editors of tomorrow?

Many translators are worried about being forced to become post-editors, falling victims of the seemingly unstoppable process that drives down quality and worsens working conditions to save on increasingly casualised professional services while reducing turnaround times. Now that nobody in professional translation would dream of working in technical and specialised domains without CAT tools, MT and PE are arguably the greatest source of anxiety among professionals. But it is important to recognise that a good translator does not necessarily make a good MT post-editor: PE requires quick thinking and the fast adoption of effective error fixes, and a constant monitoring of the trade-off between effort (i.e. time spent on PE interventions) and benefits (i.e. real, noticeable improvements in the final target text). In addition, with the exception of complete/maximum PE (where a perfect final target text must be delivered), post-editors must often settle for less-than-perfect translations, e.g. if quality is not paramount but must be sufficient for information-gathering purposes – this is something that can turn out to be particularly difficult and uncomfortable for translators, who tend to be perfectionists.

Quite understandably, not all translators are inclined to work as post-editors, e.g. because they feel that their professionalism would not be recognised or that they would not perform optimally having to revise MT output of variable quality; just like some translators are more familiar with certain technical domains, but struggle in others, or they may enjoy working on their own on large projects, but hate revising and editing the work of junior colleagues. Whatever your own strengths and weaknesses, opportunities for PE services seem set to grow in the coming years, especially because one can expect an overall improvement of MT quality in an ever expanding range of language pairs and technical domains. If you are looking forward to continuing your happy career as a language professional, it seems wise to at least consider whether you might benefit from also adding PE to your portfolio of translation services. At t he end of the day, investigating this area before your clients come asking for PE services might put you in good stead to discuss the pros and cons of this activity with them, without having to accept unfair rates imposed on you for a job that you hate or, possibly even worse, losing your clients to less scrupulous competitors.


eventsLearn more about the advantages of using machine translation and performing post-editing  as a service by attending one of Federico’s live or on-demand  ProZ.com training sessions on the subject. The full course list is available here: http://www.proz.com/translator-training/trainers/1315/courses

Federico’s next live session, Maximize Your Productivity with Effective Machine Translation Post-Editing,” will take place on February 8th at 14:00 GMT. You can reserve your seat in the course by visiting the session page and clicking the “Purchase” button in the top right corner under “Course registration”.

Did you know?

It is now possible to declare post-editing as a service you provide in your ProZ.com profile. This also means that outsourcers can search the directory for language professionals who offer this service. See the announcement: http://www.proz.com/topic/294136

Do you perform machine translation post-editing as a service? Why or why not? Comment below or tweet @ProZcom

Machine translation: Cause or solution of all evils? 2

Federico Gaspari, author of this guest post series

Federico Gaspari, author of this guest post series

Meet Federico Gaspari: ProZ.com professional trainer, conference speaker, and the author of today’s guest post.

Federico is an associate professor of English linguistics and translation studies at the University for Foreigners ‘Dante Alighieri’ of Reggio Calabria, Italy. After graduating in translation studies from the University of Bologna at Forlì, he completed an MSc and a PhD in machine translation at the University of Manchester, and has held lecturing and research positions at the Universities of Manchester, Salford, Bologna at Forlì and Macerata. His teaching and research interests include translation technologies, especially those related to machine translation, post-editing and translation quality evaluation, as well as technical and specialized translation, translation theory, corpus linguistics, corpus-based translation studies and English linguistics.

This is the first post in a two-part guest blog series by Federico on machine translation and post-editing. The second post can be found here: “Post-editing: Blessing or curse for translators?


It is no surprise that machine translation (MT) is a hot topic that polarizes opinions, especially in online communities of translators, interpreters and language professionals such as ProZ.com. However, the way in which the debate on MT has developed, especially on the Internet, is quite surprising and interesting: in the last decade or so, the discussion has moved from the feasibility and viability of translation technology to its potential applications in everyday scenarios, as a tool to address real multilingual communication needs. This in itself shows that MT is making progress and perhaps it is now time for mature reflection, leaving aside prejudice and misconceptions, primarily for translators and language specialists, who certainly are more directly affected by MT than anyone else. As an MT researcher and trainer, I take a special interest in the public debate on translation technology: here I would like to comment on a selection of recent online news stories and blog posts on the topic, discussing their relevance to professional translators.

The ever-present MT howlers

One constant in the media seems to be the regular coverage of (bizarrely often sex-related) MT ‘howlers’, whereby innocent food festivals, decent bank transactions and respectable research projects fall prey to embarrassing failures in multilingual communication, invariably featuring surreal sexual imagery: prudery may be the ultimate weapon of staunch MT opponents. No language pair or text type seem to be immune from suggestive innuendos and dirty references that are unwittingly gained in machine translation.

Whatever one’s view of MT may be, and without disputing the entertainment value of these news stories, it seems legitimate to wonder how common such spectacularly saucy mistakes must actually be to make the headlines of well-established newspapers and widely read blogs, particularly vis-à-vis more dignified and reasonable (if still imperfect) uses of translation technology. If one were to uncritically accept the implied message of these recurrent news reports, then MT would appear to be the cause of (nearly) all evils; and perhaps also the reason for impossible interlingual communication and for the lack of intercultural understanding, due to the abundant porn references randomly cropping up in MT output. This negative bias is reminiscent of the die-hard criticism that has consistently denied the feasibility of MT since the 1950s, focusing attention on (often apocryphal) stupid mistakes.

Not all doom and gloom

quotes2At the opposite extreme of the public debate on MT one finds a deterministic and almost messianic view of translation technology, typically held by those who present it as the solution to (nearly) all evils of today’s interlingual communication in an increasingly globalised world – not surprisingly, this rosy picture is often painted by those who have a vested interest in translation technology being funded, or sold, or both. These MT enthusiasts stand in the long, and mostly noble, tradition of visionaries, researchers, developers and entrepreneurs who have contributed to the improvement of translation technology in small incremental steps over the last 6 decades, through many ups and downs. In my view, the current online public debate on MT is more a testimony to the gradual successes of the exponents of this pro-MT camp, than to the farsightedness of MT’s intransigent detractors.

While I certainly do not subscribe to the rather simplistic view that MT can bring an end to all evils, and I even have my own doubts that it will ever, rather more modestly, enable faultless communication across all language barriers and cultural borders, I find it quite interesting that translation technology is increasingly discussed in unprecedented contexts – one doesn’t have to scour the Internet far and wide to come across recent blog posts written by authors with various backgrounds (computer scientists, linguists, lexicographers, language teachers, industry analysts and all-round know-it-alls), where it is claimed, for example, that: MT undermines the dominance of English as the world’s lingua franca; MT makes language learning obsolete; MT is detrimental to the preservation and survival of minority languages. To my mind, these arguments concerning the sociolinguistic, educational and cultural impact of MT suggest that, for better or for worse, translation technology is playing a far-reaching role that hardly anybody would have predicted before the turn of the millennium.

What do translators think?

quotes3Online news stories and blog posts discussing these topics attract a wide range of comments, reflecting a variety of opinions. I find it striking that people from all walks of life have strong views on these issues, which are relatively new in the public debate on technological progress, and have to date received scarce attention even in the relevant scholarly literature: while a few of the authors and commentators are experts with proven credentials in the field, lay members of the public represent the majority of those who exchange views on such exciting new topics surrounding MT on the Internet. And their opinions are hardly ever balanced or nuanced, but almost invariably polarised between contemptuous criticism and idealist enthusiasm: there is very little room to critically review different points of view, let alone to discuss doubts.

Translators are becoming more involved in discussing MT-related issues, and I for one welcome their growing engagement with these topics that capture the public’s attention. While it is perfectly understandable that professional translators are among the most vocal critics of MT, they are also the ones who would benefit the most from a thorough understanding and a reasoned discussion of the pros and cons, potential and limitations of translation technology, steering clear of sensationalist drivel. The ProZ.com community has always been at the forefront of technological developments in the translation profession, offering a platform where practitioners and trainers can openly exchange their views on relevant, if at times controversial, topics. I would argue that keeping abreast of the latest trends in MT is vital not only for today’s translator training, but also, and perhaps more crucially, for professional translators’ lifelong learning and continuing professional development going forward. The ProZ.com community can play a vital role in sustaining an informed debate on the impact of MT on professional translators and in providing them with valuable training opportunities in this area.


eventsInterested in learning about the benefits of using machine translation in your work? Federico offers several ProZ.com training courses on the subject, including two upcoming live sessions:

Turn Machine Translation from Foe to Ally on January 25th at 14:00 GMT

Maximize Your Productivity with Effective Machine Translation Post-Editing on February 8th at 14:00 GMT

Stay tuned for the second part of this guest post series, “Post-editing: Blessing or curse for translators?”, to be featured soon.

Do you use machine translation professionally? Why or why not? Add a comment below or tweet your response to @ProZcom