I saw this new feature on Hack Chinese (2.0 update) no idea how it is implemented there without increasing the review burden.
Given HanziHero’s primary focus on recognition and learning characters, i suppose integrating a new production training feature directly into the main study flow it is not feasible. It could be a valuable addition to the Extra Study tab i guess:
"In Hack Chinese 1.0, we focused on “recognition” training, where you see a Chinese character and are prompted to recall its pronunciation and meaning.
In Hack Chinese 2.0, we’ve added “production” training where you see English words and are prompted to recall their Chinese pronunciations.
Production training may improve your ability to produce Chinese words when speaking or typing, and we’re excited to get feedback on how we’ve implemented it without increasing your review burden."
Talking about my personal experience, it is true that i feel my ability to recognize characters / words much better than my ability to recall the chinese pronunciations from english words. What do you think?
Personally, I’d love for something like this too, BUT the difficulty is that it may be kind of vague to recall a specific word such as “resemble” because there are probably 4-5 ways to say that which are covered in hanzihero. Unless it were a catch-all type of solution where if you get one of the five then you’re good. But I can see how it’d already be difficult to do a reverse mapping.
Love the idea, but realistically, context matters a lot in Chinese.
I was just thinking about this idea the other day. I personally would love this as it would strengthen the knowledge of the words, even if it has limitations due to synonyms.
A potential workaround to the synonym issue specifically is to reveal a unique character (this likely gives away too much of the answer), or possibly the initial pinyin letter(s) in order to allow only one answer.
Yeah, I have no idea how they implement it on this other platform. It would be great if someone had tested it, because I assume they found a way to make it work
