Being a professional translator today has little to do with what it used to be. I am not talking about replacing all those five-thousand-page paper dictionaries with online resources: regardless of how much paper you still keep at your home office, the different tasks average language professionals must deal with on their everyday routine have changed substantially in the last few years.
As a result of the key role we play in the global economy – we are often perceived as the invisible cogs that make everything else work-, translators have joined the race where each of us must stay competitive while working faster, more efficiently, better.
How does this impact our daily work? Using standard business jargon, we are asked to deliver. And delivering means nowadays:
Improving the quality of our translations consistently, across projects, clients, and fields of specialization.
Adapting to new technologies and processes They ask, you implement and upgrade yourself.
Workingfaster. No big surprise.
And working smarter, which isn’t necessarily a nice word which actually means working for less; think of it as helping your client make the most out of each cent invested in your work. It implies all three points mentioned above, but a lot more too.
Embracing Technology… only when it suits your needs
We translators are a lucky lot. In the last few years, technology has evolved beyond all expectations, and even if most industries can benefit from it, the amount and variety of tools translators can use nowadays to boost productivity is truly unmatched.
Choose Your Weapons Wisely
We have a myriad of tools that can help us improve our daily tasks; however, depending on what we do for a living –legal translation, technical translation, localization, audio-visual translation, transcription–, we will find that certain pieces of software have the potential to become essential allies, while other tools, even if widely popular, will prove utterly useless.
The secret? There is no magic recipe, but you must weigh down those repetitive tasks you do often. Those that make you think “I could use a hand here.” Be ready to test, install and uninstall and, above all, listen to your colleagues (especially the tech-savvy ones) and learn which tools they use to spare some strokes every day.
You will see that the following set of tools features freeware, shareware, and full-license programs. My advice? Test free tools, check demo versions of commercial applications and think about buying a license in terms of “how many hours per month/year would this program save me?”
Here are some suggestions:
OCR and Image Text Extraction Software
Do you often face PDF files with tricky formatting or weird fonts that your average Word or CAT tool cannot extract properly? Check out some Optical Character Recognition software (OCR) like Able2Extract, PDFelement or the industry veteran OmniPage. Nowadays, many CAT Tools do the trick for the most common document formats and layouts; however, if you are a legal translator or deal with patents and research documentation, OCR is a must.
Translating text embedded in images can become a real pain, as no CAT tool is designed to extract words from schemes, pictures or clip annotations. Or you might just need to grab some words from a PDF and avoid the full extract process. In this case, Project Naptha browser plug-in comes to your rescue, and it’s completely free.
Wouldn’t you love having a software that works as a personal secretary, typing everything you say out loud and saving you thousands of keystrokes every day?
Well, that kind of magic does exist and is called Speech Recognition software! These pieces of fine software basically read your lips as you speak to your PC microphone and convert your words into readable text in any application you might be using. Here you can find some examples:
If your daily routine involves sending a lot of emails to clients and colleagues, and the texts that you translate have large portions of text in long segments, you might want to give Speech Recognition software like Nuance’s Dragon Naturally Speaking a chance. Once you are past the initial setup and start getting used to it, this fine software can save you thousands of key strokes a day!
If you do not need the full Swiss army knife Nuance offers, you may want to give a try to dictation software with Speechnotes or Dictation and test them directly on your browser.
A true lifesaver. Think about all the small things you do every day following the same sequence of clicks and moves in your PC. This small application runs in the background and can be programmed to automate recurring actions on your PC. Literally ANY action. From sending emails to confirming segments on your CAT to autotyping frequent texts and opening applications and menus of your choice, all with a single combination of keys.
You can make it as easy or as complicated as you want, and even if you are not into macro recording, you can find a lot of examples ready to be used in your PC at the developer website and fan forums all over the Internet. Did I mention it is completely free?
Test Every Feature of Your CAT Tool, then Choose and Discard
Auto Suggest, Auto Propagation, generic and custom Glossaries, automated Filters and a myriad of QA checks are there for you, but it doesn’t mean that you should use every single feature if they are but a waste of time to you.
I know for a fact that most translators leave 95% of the default settings in their CAT untouched. Forever. Regardless of whether they use them or not. Why don’t you try to modify some of these features and check if you actually work faster or not? For example, I found out that in many localization projects, and in several marketing presentations, Auto Propagation of 100% matches was a real pain as I spent more time fixing out-of-context matches than manually checking them, so I disabled the feature and earned some precious time.
You already know that you can access most of the common functions of any tool using keyboard shortcuts. Most programs allow the default key combinations to be modified by the user so you might want to assign a comfortable combination to the actions you perform frequently, such as confirming a segment, copy tags or apply Machine Translation to a source sentence. Oh, about Machine Translation…
Machine Translation: Friend or Foe?
Why is MT so controversial?
Rather than wondering “Is Machine Translation good or evil?”, we should be asking to ourselves “How can I exploit Machine Translation -any technology, for that matter- to my advantage?”
Discussing discount schemes with your project manager is always a delicate matter. A pain point, actually. The truth is, you never know beforehand whether the suggested rate will reflect your workload.
How should you tackle MTPE rates?
First, be open-minded and ask your counterpart to be open-minded too. Hand-in-hand with low-quality output originated by poorly designed MT engines, the main source of frustration while working on MT projects is agreeing on post-editing rates that do not reflect the actual workload required to fix all the wrongs that brainless machine did.
Your Project Manager knows as well as you do that neither of you can predict the actual MTPE rate for a new project. Why don’t you agree on setting up a pilot batch in order to measure the actual productivity and agree on a rate that is fair to both sides?
Regardless of whether you run a pilot project, you should ALWAYS track the time you spend on any project worth more than a few bucks. When it comes to justifying your rates you should always be able to ground your motivations on actual figures and not just subjective impressions.
Take into account that, just like for any translation project, there is a learning curve. You must consider post-editing your first 10K in a 50K project will probably take twice as longer as your last 10K. You need to learn about how the MT engine has worked, about its common mistakes, and how certain terminology has been applied consistently right… or not. As a result of this little research of yours, you should discuss rates based on the average time spent throughout the project.
Neural Machine Translation (NMT) is all the rage nowadays, providing a different approach towards interpretation of natural language that —I’m sparing you the technical explanation— often results in cleaner output (meaning less post editing) for many popular language pairs and fields of specialization. MTPE rates should be agreed according to theexpected quality of post-editing tasks: you might be asked to deliver a light MTPE job for low-end material (e.g. a wholesale e-Shop) or an impeccable marketing text for a high-end tech company. MT may work far better in certain language pairs and fields of specialization, hence it is easy to understand that each of these tasks must be compensated for using different parameters.
In a nutshell: Don’t think of MTPE in terms of black or white but as shades of grey, and be ready to assess each project individually taking into account a few basic but important variables in mind. Suggest a preliminary agreement and leave an open door to review rates based on actual field feedback and both your Project Manager and you will leave behind most of the tensions arising in this kind of projects.
Word Rates vs. Hourly-based Rates vs. Project-based Rates?
Word rate vs. Hourly-based rates are nowadays one of the most discussed issues in the translation industry, especially when applied to tasks and request that cannot be tagged as traditional translation: DTP, glossary building, subtitling, technical QA… and, of course, Machine Translation Post-Editing.
In the last few years we have witnessed a general trend towards a reduction of word rates in translation and MTPE jobs, don’t you think? Well, we need to be aware that the industry standards and the business model have changed for end clients, language service companies and, subsequently, translators. The higher quality we obtain from MT today often leads to a reduced post-editing effort.
However, we need to consider all the side tasks and skills that we actually deploy when we address such complex requests AND consider them when it comes to agreeing on a rate system that can no longer be based on hard, numerical data. In other words: we need to apply a far more complex mix of technical and abstract skills than we used to, so it is only reasonable to think that the rating system should take this fact into account.
Now that you have reached the end of this post, here’s an Easter Egg for you: we have discussed several tactics that will make you a better CAT tool user and a wiser Machine Translation post-editor. How about some tips that you can apply to ANY translation job and, most likely, many other daily tasks you perform with your PC?
Serious about productivity? Buy a second monitor.
Alternate difficult batches with quick wins even if it implies jumping across different projects.
No matter how focused you are on a task, do observe breaks. For the sake of your sight and your brain. Keep track of your time using any tool.
And remember to switch off your smartphone and close your email client while working. Checking your alerts every 5 minutes not only harms your productivity but also forces your brain to be in two places simultaneously and kills concentration, increases fatigue and makes you more likely to commit errors while translating. No matter how busy you are, checking your mail and phone 2-3 times a day should be more than enough to cover most needs.
Do you apply some of the productivity tips we have discussed today?
Do you have any magic moves you would like to tell us?
Follow us, share & comment this post on your social networks and help us spread the word on how to become more productive versions of ourselves! Remember: productivity is a long-distance race where the goal is to improve a little bit every day by implementing techniques that work for you. In order to achieve that, be ready to test, try and discard, but at the end of the day, you should always be a more efficient version of the translator you were this morning