Feature request:
I would like to test myself in recognizing the characters/words that I have learned thru audio. can you create a quiz/review where the system plays audio, and we have to type in the pinyin and/or meaning? in this case, you would not show the character until after the user answers. In cases where two words have the same pronunciation, you can have the system say you got it right, but ask you to give another meaning (ming2 name vs. bright, etc).
thank you! great web app!
This is an interesting idea!
I think it can work for words but characters may not work. For example, the pinyin yi4 has 30 characters associated with it in the traditional course. shi4 has 23, yu4 as well
So I suppose this could be some sort of “extra game mode” section which only covers words. The question is which words to select? It could only be scoped to words recently reviewed in the last ~month or so, or all words… though I’d love to hear some opinions on which words one would want to choose 
Maybe options like
words learned in the last n days for n in {30, 7, 3, 1},HSK m word for m in {1, 2, 3, 4, 5, 6, 7-9}non-HSK wordsI think these subsets might be easier to group up front. Maybe adding in a word limit option since some of these categories could be well beyond a couple hundred words.
A long time ago I used some random website that just un-linked sounds from specific characters for all the reason Phil mentioned. Since there isn’t a nice 1:1 mapping from sound back to character, sound training can’t also be character training. But hearing shi2 vs shi3 correctly is still valuable without needing to target a specific character. If you typed in pinyin right = correct, if you didn’t = wrong. Not sure how that’d fit into HH specifically, but a listening trainer as a thing had value (although I pretty quickly transferred to full sentences as I found them much more engaging and useful).
I was about to respond with the exact same recommendation as akitak (grouping characters/words by HSK, recently learned, etc) to reduce the number of words. Great minds!
Another idea would be to say the character/word first, then use it a sentence, so that you can get context to how the word is used. Or, you could give the character, then list out a couple of words that the character is used in. Both of these would reduce the number of homonyms you would guess from.
I’d find this option extremely useful. As to the selection of characters or words, couldn’t it be done in the same way as you can select individual items when doing an extra study session? There, you can apply a filter to have only components, characters or words, but on top of that, you can still select or deselect each item individually, can’t you? In that way, you’d only practise those characters or words you’re less familiar with, or all.
I can see the point about many characters having the same sound. But, as kaysik says, just being able to practise the identification of the pintin correctly is a huge advantage. The idea of using the character or word in a sentence is also good, but I think it should be optional. For beginners like me, it would be discouraging.
Taking aobut audio, I must say I’ve got a very, very bad ear for languages. I’m not being modest or anything, I’m just realistic in a world where many people tend to pretend they’re better than they are. So maybe this won’t be of interest to many people, but I wish there was a more straightforward way to hear the different sounds together. There are several consonant sounds that sound very similar to me, and I’d like to be able to hear them side by side for as many times as I need to hear the difference. The closest way I’ve found is to go to the sounds chart in the curriculum, open the specific sound and press the “play” button. But you have to navigate back to the other sound you want to compare it with, which may be in a completely different part of the chart, and by the time you hear it you’ve almost fotgotten what the first sound was like. And it’s even worse if you want to compare three or four sounds together. Many of the explanations for the sounds that are given here aren’t too useful for me, because they rely on English, some of whose sounds are blurry to me, to begin with. Maybe making the pinyin chart interactive (so audio would play when you click on a syllable) would be a possibility.
(Edit: by the way, I suppose there’s a logical reason why you can’t use items you learnt more than one month ago in your extra study sessions. But since most sounds are learnt at the beginning, this should be changed so everybody could always practise them.)