Sometimes this isn’t the case, which is because I sometimes rush through my reviews so I can finish other things I need to do, but I’ve been making sure that I stop doing this because I know how important it is. Additionally, I’m using this vocab deck I’m using to learn readings of kanji since I went a different method of learning them. It’s proved pretty effective so far, I’ve been able to determine some word’s readings based on context and positioning, etc.
Yeah, I know. It’s a chicken and egg problem. I don’t spend a lot of time on formal grammar study, actually. I read grammar guides here and there, but I don’t study them rigorously. Rather, I use sentence decks with comprehensible input (ie, sorted based upon known vocabulary) and have been developing intuition based upon context. While this works for me, that’s partly because of my limited time for rigorous study. Your method for tackling it will likely be different.
I know you didn’t study readings when you did the intensive kanji learning. I had a lot of trouble similar to what you’re describing after doing RTK, because it does nothing for the listening. I know you said you work without subtitles, but you might want to bring Japanese subtitles on, at least at first. They’ll give you another hook with which to catch the words you do know and start to build familiarity with the words you don’t because the kanji will tell you what they mean and the audio will tell you how they’re said. Even without active study, some of the patterns will start to form in your subconscious, making the active learning easier when you get to that content.
You’re probably right here, I’ve been avoiding Japanese subs based on the fact that if English subs are up for an English show, I tend to stop listening and just focus on reading the subs. But I actually might consider doing this now that you brought this up because I do think it’ll help reinforce the vocab I’m learning and when a word comes up that I know, but I don’t hear it, I can go back and listen for it to help tune my ear. At least that’s what I’m assuming you were getting at. Thanks for the suggestion.
Honestly it takes a while, I’m not sure how long it’s supposed to take, but just keep at it consistently. And I agree with everyone else, grammar study will really help.
Once you know more grammar, you’ll realize that a lot of what you’re hearing is not actually straight vocab, but actually inflections and other grammar constructions. Every time I learn a new grammar point, I suddenly start to hear it when listening, and it suddenly makes a lot more sense. Understanding conjugations and grammar constructions will make a ton of difference. It goes far beyond the typical conjugations you see on charts, some grammar points involve long strings of syllables. Many grammar points involve several conjugations mashed together.
For example, 食べすぎなければよかったです。This is inflected with すぎる form and ば conditional form, along with いい in past tense polite conjugation and when used together, means “I wish I didn’t ‘verb’ so much”. You could know the word 食べる, but if you don’t know the grammar points being used, you might not even be able to even make out that the base word is 食べる.
I agree. That’s the one thing that helps me with TV Japan. They have Japanese subtitles on most programs, and that is a huge help to me. Not only can I see what they are saying, I can practice reading, which is slow but steady.
Thank you for the suggestion to check out supernative! Already loving it, and it makes active listening so much more engaging and it actually goes by really fast! I was doing it for what felt like 30 minutes and it turned out to be an hour. Just curious, since I’m assuming you probably have done it, what’s your score on the listen + recall part I’d be curious to know how mine compares even though I know mine is pretty low at 1750.
@sporadic Also want to thank you for the suggestion to try with Japanese subs, I’m able to hear more words now because of it. I also decided to check out grammar again, which I originally was going to wait another month or two, but turns out I know all the vocab that Tae Kim uses in the beginning so it’s easy to follow. Like you said yourself you don’t study them rigorously, and basically my plan is the same - just 15-30 minutes a day reading over all the guides because I’ll be exposed to it during active listening.
I like the idea of Supernative, but I do find the execution frustrating at times. I very frequently find myself able to understand the meaning of a long sentence, only to find that I can’t remember the one little part they want as the answer. I get tiny little increments on my rating upward, like +2 or +3 for correct answers, and then get hammered on wrong answers, like -17. I’d rather that the problems offered a roughly equal up and down movement scheme, because it just ends up being stressful. I’m currently like 2800 or so for the listen and recall thing, but I just don’t do it much for the above reasons.
Also, it’s like… it feels like it’s the same 5 shows again and again. There’s so much out there to possibly test from. I kind of hate the shouty/intense lines from dramas.
I basically never watch dramas otherwise. Wish they had some variety shows, quiz shows, etc. Real people talking, not scripted content.
Probably harder to get high quality subtitles for those. If western tv is anything to go by, live close captioning is terrible, and shows like that have a shelf life that means they don’t get more formal releases.
I had never heard of Supernative before this thread, but after trying it for the past few days, I get a similar opinion to yours. I thought it was just my lousy listening skills, but maybe it’s just a method that needs a bit more thought in the implementation.
I guess… I mean, I understand why dramas would be easier to put into the pipeline. But I have dozens of hours of shows recorded on my DVR… I’m sure people running a site like this could do the same and manually add things now and then and eventually have a pretty decent collection of them. You’d have to be able to understand what’s being said to be able to manually create a problem for it, but if the creators of the site can’t do that, that would be a bad sign haha.
To be honest, I don’t struggle with this as much, or find it as annoying, just because I’m utilizing it to hear words better. Supernative to me is just a way to stay actively engaged in my listening. I’ve also noticed I can remember the small little part they want because I play the clip over and over, and then I can hear in my head, sort of, what’s being said there. But yeah I have to agree it could be made a lot better, plus I don’t know what the difference from 1800 to 2800 is in skill level. Either way, for me at the moment I feel already it’s helping me tune my ear just slightly to help me stay focused on listening.
Actually, you really wouldn’t have to know a lot. Problem creation could be mostly automatic. Assuming you can rip the video and subtitles from published content, and use a tool like sub2srs to create the video/audio segments along with the text, doing a simple close deletion on the text is easy. If I were to build such a tool, I’d rate the questions in terms of difficulty based upon the users which get it correct using a similar algorithm to how the users get their ratings, so as more users come, the questions that get asked to people of different levels will be refined by their own responses to the auto generated content.
Their reference to chess ratings (which make both players scores change) basically implies they’re doing something like this. Your experience with only getting a few points for a right answer, but large drops for wrong also seems to suggest this. When two players of different ability play ranked chess, if the higher ranked player wins, then scores don’t change much. If there’s an upset, though, the scores change much more. Probably the questions you got wrong were ranked lower than you, so you get a big drop when you get it wrong, but get not a lot when you get it right.
Anyway, I have built Anki decks doing a procedure like this, minus the close deletion, and I had very little Japanese at that time.
My guess is not the problem of generating the content, but making sure you have the licensing for it. Educational use and no money changing hands probably falls under fair use, but it’s a pretty grey area.
I can understand wondering about this, especially for Japanese with its limited sound vocabulary.
Other than simply doing a lot of listening (and of course two-way conversation if you have the opportunity; that might be even better for it), the only things I can suggest are:
Watching material with Japanese captions turned on. It may help you pick up on patterns and sounds if you can see clearly that the characters have said X by reading the bottom. (Even if you can’t read all of the captions yet.)
Just increasing that vocabulary count. Also kind of a non-answer, since it seems like something you’re doing as part of your normal routine regardless. But the more words you can recognize, the more parts of the sentences you’re hearing (including the verbs you’re asking about) will register as discrete units of meaning. Conversely, it may be hard to pick out the "読んだ if it’s surrounded by sounds that don’t hold particular meaning and haven’t given you any other footholds to anticipate the verb.
But really it’s just a matter of exposure and vocab pick-up. Which is also just a matter of exposure and time.
Edit – I missed that you might also not be prioritizing grammar study? I’m just one more voice at this point, but since Japanese is so structurally and logically dissimilar from English, I do agree that you’ll need to do some dedicated study of its patterns, which will also help immensely with parsing what you’re hearing.
Ah yeah, I guess that was my initial point as well. Sorry. Should read more closely.
Yeah, less ‘formal’ programming is less likely to have the kind of easily mine-able content you’d want to populate such a system. So it doesn’t appear as much.
It also seems like it’s a bit of a labor of love. In which case, the content on there is probably what the site owners personally like. I wonder if they’d do a Flo-Flo like system of buying and adding content if someone asked/was willing to pay. Probably depends on the level of manual labor involve.
The reason why I haven’t prioritized grammar is due to how I learn. I just can’t learn grammar if I don’t know the vocab that is used to teach it. That said, I just started on Tae Kim’s guide since I have a lot more vocab now (all I’ll be doing is reading his guides 15-30 minutes a day). At this point, I won’t be doing any srs on it, just reading through it and doing a little practice with it. My main focus at this point is just getting as much listening as I can to help tune my ear. I’ve already started to notice a few things about grammar from just listening the short amount of time I have. Anyways, thanks for the advice I just started using the Japanese subs which are helping a lot.
(I know you’ve already got your grammar plan sorted, but just a note: most beginner grammar text books and even Tae Kim’s guide provide lists of vocab used in the chapters - so you can always prelearn the vocab or refer back to it when needed.)
Yeah I know that, it’s just that doesn’t fit well with me because I’m not learning grammar at that point. I’m learning vocab, which if I want to learn vocab I’m going to do that with my other resources. That’s why I put it off so I would get the vocab needed from my actual vocab source. I know eventually I’ll be learning vocab from other sources, but for grammar it’s just harder for me if I do it that way.
I think you can definitely learn grammar and vocab just by listening, but the effectiveness of watching anime is extremely low at your level. It would be better to listen to something where you could at least get the gist of things. Try Nihongo con Teppei for Beginners, for example.
I’m at the point where listening to everyday conversation is pretty much “easy” and I can understand almost everything from intermediate podcasts, but most anime still often goes way over my head. Mostly it’s just about specialized vocab.
I’m not neccessarily doing it to be able to understand at this point, its just for immersion. I know that it’s not super effective, but yeah I’ll be checking out some things that will be easier to understand soon since now I have a larger vocabulary.
So, I know you’ve sorted this, but you really shouldn’t worry about learning vocabulary from multiple sources. You should really welcome it. Finding vocabulary you know from one source in another is a really good way to solidify it in your brain.
One of the problems (for me at least) with SRS systems is that the human brain is really good at taking shortcuts. It doesn’t want to do more work than necessary. So it cheats. It will retain the minimum information it needs to solve a problem. That’s one of the reasons things that are easy in our quizzes aren’t always easy to recall in the real world, because whatever trigger the brain was using isn’t there in other contexts. So it’s good to change things up. Multiple sources do that for free. Immersion also, because by its nature you’re finding the general pattern rather than the specific instance.
With regards to grammar, beginning vocabulary is beginning vocabulary (as you’ve found), so most grammar sources will use a basic set of common words that you’ll either already know, even at your level, or want to know soon anyway. I know you find it difficult when the words aren’t ones you know from your other sources, but maybe do what I’ve done. When I encounter words that I need/want, I just add them to a deck I keep for ‘words encountered in the wild’. Or I unlock their vocab card early in pre-made decks that others have made. I don’t care about the order whomever made the deck chose for vocabulary. If I need/want to add a word to my lexicon now, then I do it. If it really bothers you, bookmark the article, review the vocab for a few days using whatever method you like until they feel comfortable, then go back to it. You’ve got the advantage of available study time. Use some of it for laying out a good foundation.