A recent article talked about how the Bible could be the key to translating obscure languages. The article looked at how common languages like English or Spanish might be able to be translated using online software, but more obscure languages have little data online to provide a translation.
Researchers found one of the ironic things about machine translation was that the people who need it most are the ones getting the least out of it. If the person’s native language is not common, the person is more likely to rely upon translations to find material to read. If the language is not common, machine translations will have less to rely upon to provide the person with a translation. Of course, it is important to understand even translations of the most common languages are not as accurate when performed by computer software rather than by a human.
According to the article, linguists from the University of Copenhagen, in Denmark, found that the Bible, which has been translated into over 1,500 languages, can be used for translation into even the least common languages. Anders Søgaard, a professor at the University of Copenhagen said:
The translations are extremely conservative; the verses have a completely uniform structure across the many different languages which means that we can make suitable computer models of even very small languages where we only have a couple hundred pages of biblical text.
According to the researchers, because of the structure of the Bible, it can be used to apply knowledge of a well-known language to the understanding of a lesser-known language. This includes a better understanding of words as well as grammatical structure. The model created by researchers allowed them to compare 100 languages at a time, viewing relationships between all the languages at once. The research done by the University of Copenhagen linguists is likely to be most helpful for machine translations of less-spoken languages. According to Søgarrd:
When we develop machine translation systems and search engines, we usually feed huge amounts of manually annotated texts that contain information about the function and meaning of individual words into a computer. For historical reasons, these texts have primarily been newspaper articles in English and other big languages. We do not have access to similarly annotated texts in smaller languages like Faroese, Welsh, Galician and Irish, or even major African languages like Yoruba which is spoken by 28 million people.
Søgaard and the rest of the team plan to use their research to allow under-served languages to be better served by online translation services. Unfortunately, though, even the best translation model does not provide the same level of translation accuracy as would be appreciated by a native speaker providing the translation.
While using the Bible might seem, on the surface like an ideal way to provide translations, especially considering it has been translated into so many different languages, another recent article discussed how one woman found some things were actually lost in translation when it comes to the Bible. Aviya Kushner, a native Hebrew speaker, discussed how she found what the text was saying in Hebrew was not necessarily what the text said in English. Unfortunately, part of the problem arises from how words do not always have an exact equivalent from one language to another and the structure of ancient languages not quite the same as the structure of more modern languages.
While Søgaard and his colleagues might be headed in a good direction for the translation of some more obscure languages, at Teck Language Solutions, we realize the most correct professional translations are performed by people who are native speakers of the language and who have extensive experience translating. That is exactly what you will get with our services. Next time you need something translated, and you want to know it has been done right, contact us.