As an established business with a completed website, you’ve decided to take the plunge and translate your content into several languages. You’ve heard the statistics: Only 20 percent of the world speaks and reads English as a first language. With a vast, untapped, non-English-speaking global market out there waiting for you, your business has the potential to expand by leaps and bounds. However, as you begin to research the time and costs of translation, you wonder if you should just let a machine do the translating and be done with it. You already have great content, so it should be simple to feed it into the right program and produce an accurate translation, right?
Unfortunately, no. While there are some instances where machine translation will suffice, relying solely on this technology rather than translation services could spell disaster for your company. In this article, we will examine what machine translation is, when you can and cannot rely on it, and the future of automated translation.
What Is Machine Translation And How Does It Work?
Machine translation, also called MT, is a translation that is automatically done using software programs. Sometimes, these programs are used as a fully automated process, with no human involvement at all. Other times, they are integrated as a semi-automated translation tool. A human translator completes the translated data by correcting errors in the text and localizing the content. Hence, it is appropriate for the culture of the target audience.
Deep learning and neural networks
Modern machine translation uses artificial intelligence (AI) called deep learning. This AI type mimics human neurons and is also often called neural networking. A program uses specific algorithms to learn information rapidly, sorting through millions of characters and identifying sentence strings instead of just individual words. While this technology takes massive amounts of data, cloud computing has made it possible to achieve higher accuracy than ever. This has influenced industries like advanced medicine and the stock market and has had an incredible impact on language translation.
The Internet of Things (IoT) has improved machine translation tremendously, allowing machines to pick up subtleties in grammar and expressions. Because it can quickly translate vast amounts of data, it can be a cost-effective way to start a major translation project. It is handy for content where accuracy may not be as important as personal emails, blog comments, message boards, or general content being read for personal enrichment. It is also excellent for travelers and language students. The quality is less than perfect, but its benefits for personal use are outstanding.
Machine Translation or Human Translation?
In recent years, organizations have used tools like Google Translate and Microsoft Translator. In contrast, many promising plug-ins such as Lingotech, Babble, qTranslate-X, and others have popped up to assist users in website and blog translation. These amazing tools can bridge the communication gap between you and your overseas audience. But they are also still lacking.
Machine translation has improved dramatically over the years and will continue to evolve. However, there are some serious flaws in the system. Using machine translation for your multilingual website instead of a translation agency is generally a bad idea.
Machines have no sense of humor.
Humans can pick out idioms and sarcastic remarks, but machines aren’t always able to do the same. A device may not distinguish between a figure of speech or a literal phrase. If something is meant to be humorous, machine translation could change the entire meaning, leaving your audience confused and ready to wander off online searching for content that makes sense.
Headlines, slogans, and more
Certain words translated into other languages turn out to be inappropriate or vulgar. Consider the story of Kentucky Fried Chicken. When the franchise opened in the Chinese market, they realized too late that the translation of their famous slogan “finger-lickin’ good” came out as “eat your fingers off” in Mandarin. And when Honda introduced the “Fitta” in Sweden, they didn’t figure out until later that the term was an old, vulgar slang word for a woman’s genitals.
Machines also do not know discussings certain things in different culture is inappropriates. They don’t understand double meanings, slang, or how to avoid insulting people. In short, they aren’t culturally sensitive. Translation and localization specialists (native speakers who understand the region) can help you avoid embarrassing situations. Machines don’t care whether you are embarrassed.
Keyword translation
Keywords are another area where machine translation misses the mark. Suppose the computer doesn’t recognize the words, especially if they are obscure. In that case, it will replace or skip them, which renders the sentence impossible to understand, with keywords losing their entire purpose.
Further, if your website’s content is mostly machine-generated, you risk it being removed from the index. Search engines don’t like automated material. Your rankings and traffic could be wiped out instantly, and all your hard work and SEO efforts will have been wasted. Utilizing a translation company or a professional translator is one way to avoid this unfortunate scenario.
Translation Quality
The quality of machine translation is still not up to par. Machines can’t analyze things like tone and writing style, so text tends to sound robotic and grammatically incorrect. And trying to translate into a language like Japanese or Mandarin is still notoriously difficult to do by machine without the translation coming out nearly incomprehensible.
The Future Of Machine Translation
Machine translation is part of the natural language processing industry. This industry is expected to be worth around $2.1 billion by 2024. The process by which machine translation apps learn new language patterns is ongoing and continuous. We will see steady improvement in this technology area for many years, and it will likely become an everyday part of our lives.
However, important projects should not be left to machine translation for now. Even the most advanced translation software cannot replace a professional translator’s skills to help you grow your business in the global marketplace.
Bio
Louise Taylor is the head of content for Tomedes, where she manages its blog and business translation center. Tomedes is a translation company with a worldwide client base, serving customers with translations in 90+ languages.
