So what is the best way to practice “sound” with Skritter. I only space rep the character deck. I was thinking space repping the definition deck and covering the definition, and just listening to the sound. If it is a rare homonym, to then maybe look at definition. I want to kind of tease out if I have a slow reaction time to any of my words.
Would this be the best method to do this in Scritter? Anyone test sound specifically in a different way. I realize listening in general is better. But like space repping characters this is a kind of targeted practice that might help a beginner “acquire” the language faster as a supplement.
I was thinking I have lists of around 50 words that roughly follow the Integrated Chinese Series, I was considering to manually space rep the lists, so I wouldn’t fall into so much homonym trouble, and I can still make a separate list for “problem” words. I like this because not only is it sound practice reaction time practice, it’s also indirectly targeted tone practice which may more or less be important to target for this language.
With actual words that have same initials and finals that i confuse, I keep track of separately (as I also do with antonyms and synonyms (for different reasons) - but the actual homonyms I keep track are mostly just to be aware that I’m confusing the word sometimes, with so and so word. (might come up with a clever specific target test for these).
I’m sure someone is doing something better out there for this.
I did exactly what you suggest with Japanese. There are of course some problems because the system isn’t built to do that, but it works pretty well. You don’t actually have to cover the definition, just look away from the phone; after you get used to it, you can operate it most of the time without looking at all.
Obviously, this is far from ideal and it would be very cool with a listening card, which is something we have also discussed in the team. The problem is that it would be quite a major change and that there are other things we prioritise higher at the moment. I am, as you might have guessed from my Japanese experiment, a proponent for having this type of card built-in, with some better contextual support when needed. It’s doable, but not easy to get right. One day we’ll it, hopefully!
I am working my way through the HSK lists, and Skritter has sample sentences, written and voice-recorded, at least up to HSK4. Sometimes I hide everything, (pinyin, definition) and just “fill in the blank” based on reading the sentence. This is a bit removed from what you are describing, but in terms of training your ear, if you immediately played the audio and just followed along with the written word to figure out which word(s) fill in the blank this might ultimately be more beneficial. Of course you will be getting all kinds of context clues from the written text, so again it is hardly a perfect facsimile of audio flash cards for individual words. There are probably other programs out there that can accomplish that. I’m thinking maybe Memrise.
-What can Memrise do that Skritter can’t with respect to sound cards?
-My current test kind of kills two birds with one stone - it’s important for Chinese to hear the tone AND important in any language to quickly process what the word means by audio. Drowning it with other words might help with the context but might hurt a little with the exact tone of the word. So this limited kind of single word audio test may be good as a standalone irreplaceable test at the start.
-Cloze deletion is good - I think you can benefit with cloze deletion without audio though (as long as you are listening to actual audio/podcasts). Since sentence structure is very functional/logical. If there is an ambiguous word especially a few cloze deletion sentences that caters to that word might be a perfect targeted reading practice in those cases. (but this would probably be a precursor/supplement to a heavy reading regimen)
-I might however practice speaking/memorize select very important functional grammar points/ words in a few example sentences to solidify what their function is. (which I think is probably the most tedious of tests, but also has it’s place) (which are mostly selected from the best of my journal’s for a particular point which I work with native speakers to correct)
Memrise can prompt you with an audio cue. In Skritter you would have to click the sound icon. You could take a look at this Memrise deck to see how it might work:
HSK 4 Listening (test by audio) - Level 1
I’m sure it’s not exactly the content nor the exact flashcard study settings that you are after, but if you just look at the structure of the review settings you could probably create your own cards that would be closer to what you want in terms of which words they are and how the cards are presented.
I should say that I have not been using Memrise in a long while, whereas I use Skritter daily.
I’m a little puzzled why you would want to study individual spoken words in isolation. (Unless you are being given one word commands to follow.) It seems much more likely that you will be hearing words inside of sentences in the “real world.” Given that, you might know, for example, a third tone word in all its swooping glory as it might be pronounced by a Chinese teacher in front of a classroom or by software on the web, but when the tone changes kick in because of the preceding syllable you are going to hear something that does not match the flashcard example. (You used the phrase “the exact tone of the word” but what I am getting at here is the tone is dependent on what syllables come before and after it. Also there are all the characters that become neutral tones when they are the second syllable of a character. I suppose you would be studying those compounds as separate units. Anyway, all of the preceding is why I think that it better to study sentences in order to “learn words.”
Somehow I feel that your phrasing of “quickly process what the word means by audio” does not fully capture how we comprehend spoken input. There are so many collocations, set phrases, grammatical functions, etc. at play that I think people who are fluent are really not parsing individual words but taking in larger chunks.
As to your last comment, I think it would be difficult to learn a grammatical point by memorizing sentences that contain it, but not a worthless endeavor. I think the key is that you are doing it for things which you already had a desire to express, and so the meaningfulness and the motivation is there.
It appears the Memrise deck has the sound prompt without the button. But as long as you press the Skritter button it still plays the sound
Listening to individual words has this function - if you have audio mp3 files such as integrated chinese that contain the vocab (single senteces and 2 minutes passages) - you might not understand the meaning of a certain audio but not know exactly why - you know you know the vocab in it, it’s just not “clicking” - so audio cards are like flint to get the fire started. It’s a quick target test to bring to your attention which words you thought you knew like second nature, but really didn’t - then blast it with lighter fluid with mp3 audio files. It’s definitely a supplemental test.
You can learn grammar principles by selecting from a group of say 20 sentences (which are hopefully spoon fed to you, like they pretty much are to me) and the top ten among them. And then speak space repetition the group of 10 words as one card. I don’t think you need to find too much meaning in it. Although I do do a diary and make my own sentences and have them corrected by a foreign national, I feel like this is a supplement to getting a group of 10 good sentences and saying them.
I also listen to individual individual sentences that contain the vocab, and try to transcribe the whole meaning in 2 play throughs. Integrated chinese has 1 to 2 minute paragraphs and individual sentences.So individual sentences are a de facto “active listening” daily regimen. Along with 1 to 2 minute longer passages
However, cloze deletion for reading is far less stressful - almost passive (and naturally more analytical) compared with active listening. So with audio you can have a couple amazing beautiful audio sentences for the vocab that wold probably be your “Ace tier” examples. But then add 8 (arbitrary) sentences you use reading cloze deletion with I think it chips in a good 40% of the total meaning of you intuiting a word - even if the audio for the mere 2 golden sentences contributed 60%
Note: For anyone who is not trying to get extremely fluent in a short amount of time (and doesn’t have time for the “supplementals” (or worse confuse a supplemental with a “bedrock”) - I recommend getting something like integrated chinese (which I’m not actual using but something similar) - if I was trying to learn spanish - (simply because of the single sentences/ 2 minute passage structure I’m pretty sure it has which are the bedrock - for both passive (passive listening to the longer file) and active listening (active listen to both the longer file, (but try to catch everything) and the shorter sentences training)…if i was trying to learn spanish - I would get FSI or Pimsleur. (Perhaps also for Chinese)