It’s a different story when it comes to informal phrases: it showed a 72% accuracy when converting English casual texts into other languages. One thing to keep in mind about the accuracy of Google Translate is that it works best when translating literary text into English. That means machine translation worked better than they expected. Results from our own study on the state of machine translation for website translation showed that 10 out of 14 translation editors were positively surprised by the quality of the translation they were shown. Sometimes, Google Translate’s precision is shockingly good. For instance, since Spanish is one of its most popular languages, its translation accuracy is typically over 90%. While Google Translate is available in more than 130 languages-making it a translation tool with the broadest range of support-it also varies in terms of accuracy rate. By leaving out the middleman, it worked faster, more efficiently, and most importantly, more precisely. That means it translated from French to Japanese instead of French to English, then to Japanese. It instead directly translated between two languages. With this new learning system, Google Translate stopped using English as a go-between for translating any language. The change was massive: it cut translation errors by more than 55%-85% across many major language pairs. The result? Translations that were much more faithful, even factoring in slang and colloquialisms. Instead of translating each word, it looked at the meaning of the entire sentence. The move made leaps and bounds in its algorithm and changed its approach to translation. 10 years later, in 2016, the company developed its own framework, Google neural machine translation technology (GNMT). Google knew they needed to swap machine translation (MT) technologies to improve its accuracy. But it began to offer odd translations for longer, complex sentences. Since it translated individual words, it worked most efficiently for brief phrases. But, it soon became clear that this wasn’t going to work in the long term. When it first launched in 2006, it used statistical machine translation to provide instant translated text. Apart from that, it relies on various digital resources and common translations for languages. Ever wondered how Google Translate built its translation database? It’s heavily based on the Europarl Corpus, which is a collection of documents from European Parliament procedures that humans translated.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |