THE MACHINE TRANSLATION REVOLUTION

traducción automática

Although it seems very recent, machine translation has been around since the mid-20th century.  We have to go back to the 1950s to find the first experiments on this subject. It was a “word-for-word” translation of several sentences from English into Russian. The outcome wasn’t very successful, as the literal translation does not effectively convey the message. But it did plant the seeds of research in this field, and various research models and methodologies have emerged since then.

It wasn’t until the 1990s that machine translation would change radically with the arrival of the Internet, as the amount of online resources and texts contributed greatly to its research.

But what is machine translation?

Inteligencia artificial

By now, everyone has heard about artificial intelligence (AI), which is so abruptly entering our lives. We can already see it everywhere, and the world of translation was not going to escape it. Here’s how AI is influencing the translation industry.

Let’s start by explaining what machine translation is.  It is the conversion of a text from one source language to another target language without human intervention, that is, carried out entirely by a computer programme. There are different machine translation methods: rule-based translation and statistical machine translation, the latter being the most developed and widely used. In this type of translation, the software learns from a set of texts, called a corpus, in a certain language, that is as large as possible together with another corpus of the same size containing translations between the original or source language and the target language.

Nowadays we can find various machine translation resources, as its use is widespread. For example, websites such as DeepL or even Google Translate, provide roughly accurate translations, depending on the language we choose (they are not all as well developed).
It is also clear that they have improved considerably in certain language combinations. These are just a couple of examples, but the vast majority of large companies such as Amazon or Microsoft have also developed machine translation tools.

Benefits of AI-based translation

The widespread use of this type of translation brings to light the number of benefits that it has for both professional translators and translation agencies.

The main benefit is reducing the time required for a translation to be solely carried out by a professional translator. In other words, if we use a ready-made machine translation, the translation deadlines are considerably reduced, as in this case, the professional translator’s main task will be the post-editing of the text and its subsequent professional review.

This type of tool makes it easier to manage several translation projects simultaneously because it makes the work much easier. What’s more, it helps with the management, creation and maintenance of translation memories, which are also an essential tool for the daily work of a professional translator. Translation memories store previous translations that can be recovered in future projects. Adding all of these benefits together results in lower translation costs for the client.

Problems with machine translation. The role of post-editing

We must consider whether machine translation is 100% reliable. The answer is no. Machine translation is a very useful tool for professional translators, but we cannot completely trust its results, as communication is very complex and contains many nuances that, at the moment, machines are not able to understand. This is why there is a need to post-edit and localise the text.

When we talk about post-editing, we are referring to the review of the text that has already been machine translated so that it meets the quality level we are looking for, thus detecting possible errors in translation, grammar, spelling, vocabulary, style or register. On the other hand, localisation is the adaptation of the text to a certain geographical area, taking into account cultural or regional aspects. For example, translating into Castilian Spanish is not the same as translating into Latin American Spanish. Localisation is widely used in the translation of software or video games, but it is also very useful for advertising.

As mentioned above, the translator starts with a ready-made machine translation and thoroughly reviews it (post-editing) in order to deliver the highest quality final text. This simplifies the tasks and therefore makes the translation process cheaper and reduces delivery times.

Given all of the above, post-editing is not only our field’s present but also its future and at LexGo Translations we are already on board. Reduce time and costs with our professional post-editing services! Request a free quote here.

And we can’t say goodbye without saying thank you for reading 😉

No Comments

Post A Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Si continuas utilizando este sitio aceptas el uso de cookies. Más información

Los ajustes de cookies de esta web están configurados para "permitir cookies" y así ofrecerte la mejor experiencia de navegación posible. Si sigues utilizando esta web sin cambiar tus ajustes de cookies o haces clic en "Aceptar" estarás dando tu consentimiento a esto.

Cerrar