![]() The models that DeepL uses today can learn and abstract to such an extent that it allows them to follow a natural model and not just translate sequentially. It is where AI-based methods using neural networks show their advantage. Only then should you focus on the individual words and try to finish the translation, revisiting the original sentence from time to time to check the separate terms. It may require completely different grammar, sentence sharing, or many other linguistic methods. When translating yourself and not wanting the translation to be literal and word for word, you first need to grasp the general sense of the sentence and think about how to translate this meaning into another language. The main difficulty in creating translation programs lies not in the different meanings of individual words but in controlling the entire translation process. Old solutions, such as statistical machine translation, have just begun to give way to solutions based on neural networks, thanks to which the opportunities have evened out. DeepL entered the market when the technological landscape of machine translation was at a turning point. They allow you to use an optimized web translator, integrate DeepL algorithms with your translation software, and build new applications and services using the DeepL API. The paid version allows subscribers to unlock the full capabilities of DeepL’s AI translation technology. The free version is offered from August 2017, and the commercial version (DeepL Pro) from March 2018. Still, the innovative extension of the basic capabilities of neural networks has resulted in an unprecedented improvement in translation quality. The system uses the existing Linguee translation search engine data set. No /english-arabic, though.Jarosław Kutyłowski started work on the first version of Deepl Translator at Linguee, where he has worked since 2012. Įdit: It took me a while to figure out how to select a different language on (Their UI seems broken using mobile Firefox Preview.) Appending /english-chinese and /english-japanese to the URL shows that they already support those two, and the alignment of translations appears reasonable to me. So before DeepL starts supporting any given language pair, you should expect it to appear in Linguee first. Crucially, this depends on an existing ecosystem of bilingual websites with high-quality human translations.Īccording to DeepL's website they're a spin-off of Linguee, who provide a search service for exactly that kind of parallel data. Basically, crawl the same site multiple times with different Accept-Language headers and try to align the results. ![]() Unless they have a pile of cash to pay professional translators to produce a parallel corpus, the alternative is to scrape translations from the internet. The bigger problem is likely to be lack of training data. ![]() For example, the lack of spaces separating words in Chinese and Japanese can be handled by the same word-piece segmentation they need for German compound words. Yes, the languages you listed have some edge cases, but so do languages they already support fine. I'd expect their machine translation model to be sufficiently general to support basically any language given enough training data. ![]() It depends on which part of their approch you focus on. I've heard that the SpanishEnglish translations aren't too bad. The difficulty is that they're interested not just in word-level meaning/translation-accuracy, but also phrase- and clause-level accuracy, and those are really large (i.e. I imagine they have some linguists on the project, but they're likely to be more NLP-type linguists. Perhaps professional translators rather than linguists. > Considering Translate is such an important product, I can't fathom why they just don't hire a single linguist (or just anyone who isn't completely clueless, really) per language to register decent translations, or at least import them from a real dictionary. How many people on Google Translate actually are able to reasonably verify translations? Relatively few - and those qualified few who might poke at Translate out of curiosity are just as likely not to feel inclined to offer free labour to Google. Well, the more general problem about asking people visiting Google Translate to verify if a translation from A->B is correct is that generally people visiting Google Translate didn't know how to translate A->B or weren't very sure.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |