First, there is Statistical Machine Translation, or SMT. The best example of statistical translation is Google Translate. SMTs such as Google Translate use statistical models that draw on a large amount of bilingual text that is provided to them. Through this, they try to find statistical matches for how many times something has been translated as something. For example, if the word “apple” has been translated as “manzana” in Spanish enough times, then that becomes the SMT’s pick for the word apple. It does this for every word in the source and target language before it finally outputs a translation. SMTs can be useful for translating basic words and phrases. However, for more complex sentences, SMTs are not a good pick because they do not factor in context. This means that you can end up with some very strange sentences that do not properly reflect what is being said in the source sentence.

Next, there is the Rule-Based Machine Translation, or RBMT. As the name implies, with RBMT, the machine is translating based on the rules of grammar that it is given. It goes over the source language to analyze its grammar and then looks through the target language to see what kind of grammar rules need to be followed when translating the sentence. After it is done, it translates the sentence according to the rules that it looked up. While this might sound as if there might be fewer problems than SMTs, RBMT translations still need a great deal of proofreading to ensure that there are no problems with the text.

Third, there is the Hybrid Machine Translation, or HMT. The HMT takes elements of RBMTs and SMTs and then uses a translation memory. The addition of a translation memory makes the overall quality of the translation higher than that of SMTs and RBMTs, but there are still problems with using a HMT. For one, it still needs extensive editing, meaning that human translators will be needed to review all the translations.

The fourth, and final type of translation is Neural Machine Translation, or NMT. This type of machine translation uses neural network models that are based on the human brain to make statistical models for translation. When they work, NMTs can provide some of the most accurate translations that would need little to no editing potentially. One of NMT examples is DeepL.

Overall machine translation can be used as a tool to expedite translation, but whether it can produce clear and accurate translation for high level discussions, complex concepts, nuances and context that is still in question and requires time to advance it. This once again proves that the value of professional human translators is not replaceable just yet!

For the best results when translating, it is important to always think about the context in which the source text was written and how it might be perceived if brought to the target culture. What this means is that language and culture are always very tightly intertwined with one another, so what one language might find normal, another language might find abnormal or outright offensive. For example, in the Hopi Indian language, there is no word for time, so their understanding of time is very different from most of the world where there is a defined word and definition for time. Similarly, divorce does not have its own native word in most Indian languages, and they instead use the English word “divorce,” if they ever want to get a divorce. However, divorce rates are much lower than they might be in other places in the world, due to their culture seeing marriage as a lifetime commitment. These kinds of cultural differences is why you have to do a little research when you are translating.

Similarly, in Japan, they have adapted a lot of English words into their own language and use them on a daily basis, but if an English speaking person heard these words, they might have a different idea of what they mean. For example the word “lamp” might mean a kind of light that you might find on your desk or in the room, but in Japanese, it is used almost exclusively for indicator lights. So when you translate languages that use words that appear in the target language as well, be sure to make sure that they have the same meaning in both languages.

Another thing you should be careful of is avoiding idioms altogether if you can help it. Idioms are commonplace in any language, but there will not always be an equivalent idiom in the target language. Unless the idiom is very essential to the translation, you should try to figure out a different way of phrasing it. For example, the phrase “two birds, one stone.” There may be equivalents to this in some languages, but if the target language does not have an equivalent, then you could instead say “accomplish two things with one action” or something of the sort.

While translation can be a fairly straightforward process of just taking the source words and then convert it into the target language, it can often be much more complex than that. While fields such as technical and patent translation might encourage a more one to one translation, there are some styles, phrases, and references that just won’t translate if you adapt a one to one style. That is where some creative liberties come into play where the translator tries to find replacement words or phrases that won’t always have the exact meaning of the original, but will be able to give the same nuance or elicit the same kind of reaction as the original did in the source language.

What is hyperlocalization? Hyperlocalization aims to identify, understand, and target unique micro-segments and to then localize product/service, communications, and business models in order to deliver customized solutions to meet hyper-local needs. In essence, hyperlocalization is a super focused localization process that aims to make sure that they are meeting the needs of a specific demographic by making hyperlocal content. Hyperlocal content is targeted to country-specific or locale-specific and even global-specific micro-segments within specific demographic, geographic, psychographic, behavioral, cultural or other segmentation groups. Below are some examples of hyperlocalization.

For hyperlocalization based on geographies, you might see companies changing how they localize content for a specific region or country so that it is specific to what users for that area might expect. On the other hand, hyperlocalization based on demographics focus much more on finding micro segments based on unique demographic combinations or emergent micro-trends related to age, gender, income, or education. So focusing on specific groups like millenial foodies or centennials craving brand authenticity and trying to appeal to them through targeted language and advertisements is how you might get their business.

So why is hyperlocalization so important to understand and know how to implement? Well, when localizing something, even though it is different than translation in the sense that you are tailoring the words you use to reflect the audience you are translating for, still it will be more of a shotgun approach than a targeted approach. It might sound perfectly and linguistically alright in the area you are localizing for, but that will not matter if you are not making sure you reach the group of people that would make up your core audience or consumers. Hyperlocalization achieves this by making sure that you are specifically focusing your efforts on the people who will have most interaction with the product you are localizing.

With hyperlocalization, you are getting the maximum amount of connectivity with the audience that you are trying to target with. Also, just like with localization, what you are localizing sounds natural to the target audience. The only difference is that you are making as much of the content that sound pertinent to them as possible. Hyperlocalization may not be something that you will be expected to do in the localization industry, but it is something that you should know about, so you know what is expected of you when you are asked to do it.