If you’re using CantoDict, you’ll know that some very nice people have taken the time to record sound samples in mp3 format. You might not know however that you can copy those sound files to your computer for use in Anki (or SRS of your choice). This means that you can quickly create flashcards that simulate being read to – which is good for all kinds of things (pronunciation, reading, listening, popularity, cooking skills…)
In CantoDict, if you right-click the sound icon next to a sentence and hit “save link as”, you can save an mp3 to your hard drive. If you save it to your Cantonese deck’s media folder, you can then use it in Anki; just press f4 in an empty field to add it to a card. You can then copy and paste the accompanying characters and translation into separate fields.
There’s also an Anki plugin called “Basic Chinese Support” – if you download that and select the “Cantonese” model, it will give you the Jyutping for any characters you put in (if there’s a choice, you just have to double-click the correct pronunciation).
Anyway, here are a couple of screencasts of adding a new card and reviewing one; in the review, the audio goes with the question rather than the answer, and the goal is to understand what’s being said and repeat back the sound as accurately as possible. (Click to enlarge:)
This might also work for other languages too: if there’s a ready source of mp3 samples available then it should be easy. If there are samples but you can’t readily download them, you could also try using a soundcard capturer and/or audio splicer to create your own. If you’re just downloading mp3s, each card should only take ten or fifteen seconds to add, so it’s reasonably quick.
Hope that’s clear – if not (or you run into any problems), let me know!
 The media folder might be hidden – if (in Anki) you go File>Open, it’ll show you where it’s stored.