Like all professionals, translators need a range of tools to carry out their work, from the most generic to the highly specific. While not all translators work in the same way, there are a number of points in common from one professional to another. However, many of the essential tools are not generally known to the general public. Here is a selection of the tools most commonly used in our industry:
It may seem rather obvious, but today’s translators work almost exclusively with IT tools. The days of the typewriter are long gone, and the days of pen and paper even more so. Most translations are done using word processing, spreadsheets or slide presentations. It’s a bit like a translator’s ABCs, and the office suite really is indispensable.
While everyone is familiar with the Microsoft Office suite, it is not the only one on the market. There are other solutions, some of which are free and open source, such as Open Office or Libre Office, which are even favoured by some institutions.
Whatever solution is used, the translator must know how to use it to the full. Mastering the basics of word processing software is not enough. The nature of the job requires them to use essential advanced functions on a daily basis. Comparison, change tracking, parallel view of two texts, macros… are all essential.
Let’s make no mistake: a translator cannot master the entire vocabulary of the languages he or she works with. There is a long list of little-used terms, outdated terms, terms that are very specific to professional jargon or terms that simply require clarification.
However, a native speaker only uses between 3,000 and 5,000 words, of which only a few thousand are used in everyday discussions. A dictionary like the Petit Larousse illustré, which is far from being the most comprehensive, already has over 35,000 entries. Multiply that number by at least 2, since we’re talking about source and target languages, and you’ve got enough to forget a few words here and there! In short, it’s only natural that translators sometimes need to dig through a dictionary.
Beyond clarification, synonym dictionaries are a precious asset. They can be used to make up for a lack of inspiration, to find an exact nuance or sometimes to think of another approach to an inelegant sentence.
Although linguists are still very attached to paper, the need for speed in their profession inevitably leads them to turn to IT. There is an abundance of information available, especially online: Larousse, Robert, Wordreference, Urban Dictionary and Synonymes.com are just a few examples.
In addition to dictionaries, translators often work with glossaries, which can be either monolingual or bilingual. These are often provided by the customer and are used either to clarify acronyms or neologisms, for example, or to standardise vocabulary in line with an editorial line or brand image. For example, I worked for a brand that asked us to use the term “soulier” and banish the word “chaussure”.
The same problem can arise with in-house jargon. Personal titles and department names are good examples: from one company to another, it’s easy to find different names for identical functions.
Glossaries can clear up many doubts and avoid having to go back and forth with the customer. They also help to avoid a number of pitfalls that could be seen as a quality flaw.
While most glossaries are supplied in tabular form, they can be converted into formats that make them more interactive. In particular, so that terms can be identified automatically and the translator warned when using the software dedicated to translators par excellence: CAT.
I’ve already mentioned this in the FAQ section of the site, but if there’s one indispensable tool for translators that the public doesn’t know about, it’s CAT software.
CAT stands for Traduction Assistée par Ordinateur (Computer-Assisted Translation). Now, some people will say that this is by no means a must-have. In fact, for some documents it doesn’t add much. It can even be a hindrance (literary or highly editorial texts, for example). But for the rest, it is one of the most useful tools in the translator’s palette.
While there are variations in functionality from one software package to another, the main principles always remain the same. Texts submitted to the software are broken down into segments (more or less sentences) which are then compared with translation memories. In simple terms, translation memories are databases containing all the translations already carried out in previous projects. The aim is to retrieve similar sentences and their translations in order to save time and maintain a certain degree of consistency.
But the benefits of this software don’t stop there. They also provide a plethora of other tools: interactive glossaries, spellcheckers, automatic number localisation, unit of measurement conversions, quality checks, spellcheckers… and even, in some cases, automatic translation engines.
Yes, linguists are language lovers and sometimes a little rigid about the rules. But they’re still human, so they can make mistakes, typos and other blunders. Not that they don’t know the rules, but caught up in the heat of the moment, with that sentence rephrased 15 times… well, there it is, the mistake that was bound to happen. Fortunately, the experienced translator is equipped with a powerful spellchecker.
Everyone is familiar with the spellchecker found in the Microsoft Office suite, especially Word. But there are far more powerful tools, some of which are integrated directly into certain CAT software packages, and some of which are completely independent (Antidote, for example). These tools not only provide the correct spelling of words, they also check grammar, agreement, syntax and, in some cases, even give style advice!
But you need to choose the tool you want carefully, because these are really based on the languages checked. Grammarly, for example, which is getting a lot of publicity at the moment, only deals with English.
Automatic translation engines
I can already feel the purists sounding the alarm… But yes, apart from the usual small village of diehards, translators are also using machine translation. There are several ways in which this is reflected in everyday life.
Firstly, on a more or less ad hoc basis, as and when required. A sentence that’s a bit convoluted, a need for ideas, an unclear word, the need for a quick synonym… Translators often look at what the machine suggests. There’s nothing wrong with that, since it’s just one source of inspiration among many.
Systematically in certain translation processes: augmented translation, post-editing. Here, the translation provided by the machine serves as a complete base. The translator will then rework the automatic suggestions according to the objective and the quality provided by the machine. This can save a considerable amount of time, but it can also be a real headache and take longer than a human translation. It all depends on the type of text.
It doesn’t matter what the translator is using the machine for. It is important to choose the right engine to provide the basis for the work. First, there is the problem of confidentiality (see Translate.com’s setbacks), then quality. Not all translation engines are created equal, and the same engine is not necessarily equally effective in all fields and for all language pairs. Numerous solutions exist and new ones appear regularly. In any case, it’s a subject that fascinates me and I’ll be talking more about it in other posts.
Another acronym! OCR stands for Optical Character Recognition. The principle is quite simple: you ‘show’ this programme letters, words, sentences or texts and it returns the same thing transcribed into an editable format (.txt, .doc/.docx or other). What’s the point? The aim is to make editable what is not originally editable. This mainly involves images (whether handwritten texts or scanned or photographed pages), but also PDF files, which are often uneditable. This allows the translator to do many things:
- replace the existing text with the translation
- recover a specific page layout without spending hours recreating it (assuming this is possible);
- quickly find out the number of words in a text, in order to estimate the cost of the translation;
- import the text into a CAT tool;
- use machine translation…
In short, there are many applications. But as with machine translation, you need to choose the right service from the plethora of solutions on the market. Here too, confidentiality is a concern, and quality can vary from one software package to another, as well as from one language to another. The Latin alphabet is not recognised in the same way as Cyrillic or Chinese characters… I’m simplifying, because the reality is much more complex, but I’ll talk more about this in another article.
If you’re a translator specialising in the audiovisual sector, you’ve just added another tool to your list. Subtitling requires a number of specialised tools.
First of all, those used to transcribe the spoken text in the source files with time codes, which will later be used to say “display this subtitle at such and such a time”. We therefore create a sort of first subtitle in the source language. There are various ways of helping the translator with this task, with varying degrees of automatism depending on the solution. Transcription of audio and video texts is one of the major areas of research in automatic natural language processing, in the same way as translation, but I’ll come back to this another time.
Then comes the actual translation. The translation must be directly visible in the context of the video. This avoids any problems of text length or subtitles running too quickly. Finally, other tools may be needed to inlay the subtitles or to convert the formats of the final files.
Translators are not graphic designers. However, they do have to juggle with a multitude of clients and therefore a number of graphic formats. The majority of documents to be translated are published, so they have been laid out and sometimes carefully thought out. The translator must be able to follow through and produce work that is just as clean as the original document. From simple layouts to entire page layouts, including all image formats, the translator must be able to navigate all types of file.
It’s not enough to simply replace the text, you also have to make sure that it fits into the boxes, which is not always easy. This is not always easy, because text does not take up the same space from one language to another. Because the words are sometimes longer, simple words in one language require a complete sentence to convey the same meaning in another, and so on. This is what we call “proliferation”. French is on average 20% longer than English. The slightest layout will inevitably require checking after translation, if only to ensure that the text does not slip off the page.
Just one more acronym! TMS stands for Translation Management System. As its name suggests, it is used to manage all translation files and projects. It’s a bit like a translator’s ERP, enabling them to keep track of their files and their progress.
Here too, there is a whole range of more or less sophisticated products with a variety of functions. Sometimes this tool is coupled with the CAT tool, sometimes not. To be honest, for a freelance translator it is not necessarily essential. A simple Excel spreadsheet can often suffice and prove quicker and more practical on a day-to-day basis.
On the other hand, translators need to know how things work if they work with translation agencies. Many translation agencies use these TMSs. It is up to the translator to connect to it to retrieve the work to be done and to return it.
This is not a specific tool, but rather a broad category. A translator is a business in his own right. To a greater or lesser extent, they need all the things that keep a business going.
Translators have to keep track of their finances and accounting (quotes, invoicing, VAT, payment methods, etc.).
They are also salespeople, because they have to find their customers. CRM and other sales tools are therefore part of the toolbox.
He’s also a marketer/designer/informatician… because nobody can do without a presence on the net these days.
In short, every freelancer wears many hats, each of which may require additional tools.
What are your favourite tools?