Indicate desired kanji via audio

I think altering the initial auto-pronunciation to {word} の {kanji}    e.g.  「べんり」の「り」   would be much more helpful than just sounding the word itself.

Reason being:

When words with multiple kanji come up, you can't tell from the initial audio which part is being taught / asked for.

I think this is helping to cause false associations - e.g. having to write 便 or 利 after hearing べんり makes the two kanjis' readings more difficult to keep separate and therefore easier to confuse.It also negates many of the advantages of the audio when you always need to look at the underlined hiragana anyway.

See https://japanese.stackexchange.com/questions/69798/how-to-address-refer-to-a-specific-kanji-in-speech/69800#69800 for the explanation and https://m.youtube.com/watch?v=PhtOewdIxII&t=105 and https://m.youtube.com/watch?t=255&v=sJNxPRBvRQg to see this method in use.

No need to do anything with individual 'on press' pronunciations, since the user has already seen the hiragana by then.

Thanks.