Apparently deepl thinks that the discovery of America cannot have taken place in 1492 and autocorrects the year to 1942!
Is that the director of Harry Potter?
Yeah. Also, Iām increasingly noticing that when translating larger blocks of text, Deepl will occasionally replace a sentenceās translation with the translation of the sentence next to it, and itāll keep using that translation even when Iāve deleted the adjacent sentence.
For example:
Deepl, only one of those bullet points actually reads āDo not use if you have scars, rashes, or other skin problems.ā And if I delete that bullet point from the original text, it simply removes one of the āDo not useā¦ā lines from the translation.
Itās doing me a concern.
Off topic. Iāve never used deepl, how does it compare to Googleās translator?
EDIT: Iāve noticed now that it doesnāt allow translation from English to Japanese, only the contrary, so it canāt fully replace Google as of now
Toss all the numbers in a bag, shake it up, and draw them out at random.
Itās several orders of magnitude better, with the biggest caveat being that it tends to favour a natural-sounding translation over an accurate one. So, sometimes it⦠guesses.
Ohhh, I noticed that too! So Iām not insane!
The trick is that you need to set the āfromā language to English first. Then Japanese appears in the ātoā language list.
This could be that they trained the network on texts that contained mistakes like that, and so now the network has learned the mistake!
Thanks, thatās very useful to know.
1942 was the battle of Midway so maybe it just parsed a lot of text containing that year.
its not a translater, itās a fancy statistical language model
Yeah but thatās pretty much all of us as well. lol
lets not repeat the AGI debate we had before (fun as it was)
I like that it goes for a more natural-sounding translation over a literal one, but Google has the romaji along the bottom which is really nice if I am copy/pasting a chunk of text I canāt read. Itās not perfect, but it is nice.
Perhaps deepl canāt decide between Japanese and Chinese - so it must be German
However it works here. Does deepl know that I am using it mostly for Japanese?
This is the first time Ive ever had issue with deepl auto detecting Japanese, so it surprised me.
I think that is related to kanji-only expressions, where there may be ambiguities with Chinese. I am using google translate as ābackground translatorā with bookwalker, and I have often been told ātranslated from Chineseā when I mark only kanji. It doesnāt give pronunciation then, but gives the correct translation. When I select more than just kanji, it recognizes Japanese and all is well.
ę²³č± was a word in an ancient Germanic language, which later split into several new languages, including German and English. The word ę²³č± managed to remain unchanged in both of them