During the Mini Stories in particular I’ve come across multiple Furigana that are using incorrect translations from the audio… It would be a nice feature to edit the Furigana. I know maybe in the future I wouldn’t need Furigana if I was more advanced but right now as it stands I think it would slow me down too much to individually click on every Kanji.
AFAIK, there’s no way to directly change the furigana. You may be able to indirectly change it by editing the sentence to combine kanjis so that the “parser” can interpret it as “one word” instead of two separate requests for two separate words. which may also change the furigana.
With that said, you can’t modify the LingQ curated lessons (at least to my knowledge.) In this case, you could download the audio, copy the text from the lesson, and then reimport it. With that work, you may still have incorrect kana.
ふりがなを日本語で編集するなら、文章の編集で直していくしかないと思います。これは、私が学んでいるフランス語でも起こっている現象です。
日本語の読みは同じ漢字に対して複数あります。
当たり前に使用している私でさえ、これを初めから学ぶ方達はイライラすることと思います。
オノマトペも多いですしね。
AIでの限界をヒシヒシと感じるところです。
Thanks for the reply! short of editing the text in the audio file, I think I’ve found a workaround. there’s a setting to only use Furigana on status 1-3 words… therefore, if I come across a word with incorrect audio ill try my best to learn it, then I can change it to “known” so the I don’t see the incorrect Furigana.
I think some of the words bug out and still show the furigana even after you mark them known. Not too often though.
I hope they fix this. They should both improve the AI that does the furigana and also make it use editable. This would make LingQ amazing for Japanese learners because it would solve the huge problem learners have with Japanese - the writing system. I’m learning Korean right now and I can say that using a regular alphabet instead really simplifies things.
+1 to the feature request to be able to edit the auto-generated furigana. As a beginner, it really throws me off especially when I study the flashcards as they don’t match what the original audio sounds like.
It’s my third day on LingQ learning Japanese. The system is very promising, and I particularly like the way that words can be joined arbitrarily to form phrases so that I can practice longer sound sequences to train my ear. However, I’ve found something annoying: the text-to-speech does not work properly. In the first mini stories, the Japanese words change their pronunciation when I combine them into phrases, e.g. 七 as nana when it is an isolated word, and become shiti when in 七時. Is it the rule in Japanese? Or is there really no hard rule for that? More frustrating is that sometimes と is pronounced as TOO (i.e. in English), and 会う is pronounced as AID-YOU. When doing the flashcards, sometimes it included additional sounds (at the beginning or at the end, it repeats the first word or last word in English or in Japanese).