I am very confused as to how the Japanese sentences are coming out. I go through the words one a time, re-read the sentence and see if I understood what it said. I click translate sentence. Sometimes I’m right and sometimes its completely different to what I searched because its overly literal. I have DeepL as my main translator with Jisho as my main dictionary. One story that comes to mind is Gon the little fox under the books category. I tired reading it regularly and then with the simplified AI, the Ai gets a gist of the story but things change, including tenses, which makes me hesitant to use it as I can’t determine the accuracy of it. Additonally, I had imported a YouTube video the other day and the romaji pronounced the Kanji differently than what it was within the video, this is confusing and likely happening for other words as well. Should I stop translating the sentences altogether to check if I’m correct and just go by it word for word? Also, if there are any Native japanese speakers or speakers of japanese at a very high level I’d like to know the accuracy of the Ai. Thank you in advance.
Hey,
Nothing new, everything is known. They don’t care at all, so you have to get along with those bugs.
One solution for you :
- only use the Pop-up for translation. Take Google Translate or DeepL but don’t use the “translation by sentence”. Pop-up, at least, are working well.
- download Yomichan or 10ten to help you with the furigana, the ones from LingQ are false. If you use an iPad or Android tablet, just add the Jisho popup but turn off the furigana (i’ve learn so many wrong furigana because of that)
- leave a review to help the newcomer to know how is it going with the japanese.
Good luck on your journey.
-
Good to know the translations word by word are accurate at least. And yes, I have it set to DeepL
-
I don’t really know what you mean by this, I installed them and neither of them are working. Also, I’m using romanji not furigana unless it has a setting for that as well, Whilst I can read furigana I do find romanji much easier and becuase I know how japanese is so supposed to be pronounced I don’t run any sort of risk.
-
By this I assume you wanted to know how my Japanese was going? Well, LingQ has been helpful but parts of it have been confusing to use. I’m very new to japanese still, and I am hoping to learn it. LingQ is fun, but with the inaccurate translations at the moment its a little disheartening.
Upon figuring out how 10ten works, it’s still confusing. It’s only translating singular parts of a word rather than the full word itself. For example the word Norimashita it has given me these translations just for the one word:
To get onto (A veichle)
Lime (peru)
Right under, Directly below
Tongue
And LingQs translation for the word is “Rode”
I think I understand how this works now, thank you. But disabling furigana or romanji is gonna hinder my learning unfortunately. It’s the only way I can currently read what’s happening at a fast level, without them I’m going to be going word for word looking up how to read them. I feel doing that is going to slow me down a lot especially as I don’t know how to read Kanji nor do I want to neglect LingQ and study Kanji in the meantime as I’ll forget the vocab I have learned more than likely.
Yomichan was sunset by its owner last year. Yomitan (with a T) is its successor. If you are in Chrome you can get it from here. As they say in the chrome reviews, it’s a bit hard to set up but well and trully worth. The github repository has a basic usage guide.
Are you talking about English translation or about text-to-speech?
I found lots of problems with the “click on the word” speech in Japanese in LingQ. For example, particles like “wa, ga, na, mo” are not pronounced: instead they are spelled out in English letters! That’s right: if you click on が, the voice says “jee aey” instead of “gah”. That isn’t a mistake: that is a joke or sabotage.
The “click on a word” also has a problem with kanji. Most kanji have two or more readings, depending on the sentence. In one sentence 外 was “soto” (outside), and the “say the sentence” button correctly used “soto”. But the “click on a word” voice said “gai”, which is a different reading. The furigana was also “gai”.
I use the 10ten Chrome add-on translation dictionary. But, though it lists all the readings of a kanji, it can’t tell me which reading is correct in this sentence.
10ten is a dictionary. It does not teach you a language.
乗りました = norimashita. This is a conjugated verb, so it has many different endings, just like verbs have in French, Spanish, Russian and many other languages. The verb root is “noru” (乗る). The ending “-imashita” means past tense (polite form).
Words in other languages do not have 1-to-1 exact matches in English. They simply don’t. “Noru” can be translated as “get on board (with)”, but think of all the different meanings that can have: board an airplane; mount a horse; join a group; agree to a plan; etc. Just like English, one Japanese word is used with various meanings in different sentenes. And the more different the languages are, the more strange those those meanings might seem.
If you want 10ten to show you the meaning of norimashita (乗りました), you must hover the mouse over the first character (乗). If you hover it over a different character, you might get “rima” (Lima, Peru) or “shita” (under) or “shita” (tongue).
Written Japanese doesn’t have spaces between words, so you can’t hover the mouse over any letter in the word and have 10ten find the word. You have to hover over the first letter.