In the years when Google Translate was first available, with wrong translations He produced material for internet comedy. Those who tried to talk to strangers with translations were frustrated by the other party’s inability to understand the sentences. In fact, the “chicken translate” joke, which many of us know, even took place in movies and TV series. Over the years, Google Translate has also improved in translations, just like in other translation applications. improvements started to stand out. In fact, the translation jokes gradually began to decrease and the conflict between people due to mistranslations began to end.

Well this is what you see in Google Translate improvements how was it? where people can chat quality translations how did it originate? Let’s explore it together.

Google Translate worked with statistics when it first came out.

Google Translate first appeared in 2006. In the early years of its introduction, it was largely to online statistics was based on. system when you type a word the entire internet database It scanned it, found existing documents translated into the selected language pair, and presented the most commonly used version of a sentence as a translation.

The first working logic of translation, short sentences and single words It worked fine for him, as the translation was easy to translate single words, but the subject with complex grammar when it came to languages ​​and long texts, it had bad, funny results. In short, Google Translate was unable to produce meaningful translation results in context and sentence structure translations with its first technology.

Fundamental changes started in 2016 with neural machine translation technology.

Google said in a statement in 2016 that it had made a big improvement in web and mobile translations. this development “neural machine translation” It was based on a technology called This improvement first came about in the French, German, Spanish, Portuguese, Chinese, Japanese and Korean languages, including both Turkish and English. Translation application thanks to the technology called neural machine translation; text now in small pieces does not divide and translate, instead all sentences at once it turns into patterns. This improvement results in better syntax and more natural and meaningful results.

Moreover, this technological advancement is in translation, says Barak Turovsky, product leader of Google Translate Equivalent to 10 years of development level. Fundamental improvements to Google’s translation app have come a long way in these years.

Offline translation has also been improved by 20 percent in some languages.

The offline translation feature can be used with additional packages downloaded from the Google Translate application. Online language packs, where there is no internet access known for its use. This feature, which saves the lives of those who do not speak foreign languages ​​and do not have the Internet, is perhaps one of the most popular and widely used features of the translation application.

According to Google’s 2019 statement, Translation is available in 59 languages ​​with the innovation of the offline feature; word choice, grammar and sentence structure. by 12 percent gives accuracy. This accuracy is also available in some languages, such as Japanese, Korean, Thai, Polish and Hindi. up to 20 percent can achieve.

Translation has strengthened its power by leveraging artificial intelligence.

google, AI (Artificial Intelligence) and algorithm based It also uses learning methods. The translation app discovers millions of documents that have already been translated by humans and based on these discoveries, it can generate translations using patterns in the text. Translation over time; it contains more patterns, receives input from real people and improves translations.

We see how the results improve when we put the puzzle pieces together.

In the 13 years since its launch, Google Translate has used “neural machine translation, based on rewrite paradigms, processing on the device techniques such as: artificial intelligenceWhile these Google-developed technologies are not perfect in translation accuracy, they have brought about visible changes compared to the past. At least we no longer encounter funny and strange results in translations.

In addition to these developments, Google Translate consists of people who translate sentences and words, check that the translations in the system are correct and contribute to performance improvements in languages ​​with little data. Google Translate Community It also has a volunteer community. In addition to this community, a regular translation user also checks whether his translations are correct. by scoring It can contribute to Google Translate.

So what do you think of Google Translate’s translation results? Are you satisfied with the translations?

Sources: CNET, Google Blog, INC, TTC We Translate, Venture Beat