nest0r Wrote:@foodcubes - For films, see the ‘japanese subtitles for films’ thread; I think it might be easier to simply look at the filenames, then Google for those filenames to find the associated videos. This will turn up plenty of direct download links.

Others have mentioned d-addicts for subtitles, and likewise if you Google for the filenames you should turn up plenty of links.

By the way, I think it'd be great in the future if Anki and subs2srs are integrated into classes, but who knows if the world is ready yet.

Dang it. I finally got a Battle Angel Alita video, with English Subtitles that match, but no Japanese subtitles. I forgot that Japanese subtitles are hard to come by. (By any chance, does anyone know where to get Japanese subtitles for Battle Angel Alita, My Neighbor Totoro, or Nausicaa of the valley of the wind, or something easy for a beginner?)

Ok, so you're suggesting:

Find Available Japanese Subtitles -> Find Video (Typically Japanese Language with English Subtitles -> Create subs2srs/Anki Deck


On a related note, would it be a good idea at least to create an online repository for Japanese/English subtitles? (Your own (Nest0r's) subtitle catalog has a subtitles I couldn't even find on http://kitsunekko.net)

i think someone transcribed my neighbor totoro and nausicaa on a site somewhere.
Maybe try korean subtitles. It sounds crazy but it actually works
LINK: http://forum.koohii.com/showthread.php?p...#pid117636
well at least for this case you need a subfile anyway for sub2srs.

Thanks for uploading the ff ac decks! I appreciate it, is there a ffac_high quality version too? Just curious ^^ (sorry, diehard ff fan )

sorry to reply again, but can you explain to me how to turn these tsv/avijpgs into a anki deck? Sorry for the questions

animehunter123 Wrote:sorry to reply again, but can you explain to me how to turn these tsv/avijpgs into a anki deck? Sorry for the questions

http://subs2srs.sourceforge.net/#importing_into_anki

foodcubes Wrote:

nest0r Wrote:@foodcubes - For films, see the ‘japanese subtitles for films’ thread; I think it might be easier to simply look at the filenames, then Google for those filenames to find the associated videos. This will turn up plenty of direct download links.

Others have mentioned d-addicts for subtitles, and likewise if you Google for the filenames you should turn up plenty of links.

By the way, I think it'd be great in the future if Anki and subs2srs are integrated into classes, but who knows if the world is ready yet.

Dang it. I finally got a Battle Angel Alita video, with English Subtitles that match, but no Japanese subtitles. I forgot that Japanese subtitles are hard to come by. (By any chance, does anyone know where to get Japanese subtitles for Battle Angel Alita, My Neighbor Totoro, or Nausicaa of the valley of the wind, or something easy for a beginner?)

Ok, so you're suggesting:

Find Available Japanese Subtitles -> Find Video (Typically Japanese Language with English Subtitles -> Create subs2srs/Anki Deck


On a related note, would it be a good idea at least to create an online repository for Japanese/English subtitles? (Your own (Nest0r's) subtitle catalog has a subtitles I couldn't even find on http://kitsunekko.net)

Kitsunekko is for anime; d-addicts (subtitles section under Japanese) is for jdorama; my archive and Project Modelino and wherever else is for films. All of these (except for mine but I mentioned kloofy in that thread) should also have the English subtitles on the same site. Those don't have to be the same, though it helps, because you can tell subs2srs to use the timings from whichever subtitles you prefer.

And yes, that's what I meant! The order, I mean.

animehunter123 Wrote:Thanks for uploading the ff ac decks! I appreciate it, is there a ffac_high quality version too? Just curious ^^ (sorry, diehard ff fan )

There is, I'm using it in my deck (with the videos being full 1080p). The file is huge though, so I didn't upload it =(. I find that having the video in such high quality isn't that good an idea though, it just fills my dropbox up faster =/. Anyway, if you want the "High Quality" version, then I think you'll have to make the deck yourself, since it would take me ages and ages to upload it.

So I've just tried making my first subs2srs deck, but have run into problems.

For the movie "Crows Zero": I downloaded Nest0r's Japanese subs, then English subs (from Allsubs.com), then I got the movie. I used subs2srs to process the cards. But after trying to import into Anki, only two cards exist, both of which lack the Japanese phrase. And it looks like the subs are png files, rather than text in the .tsv file.

The video and audio files look right. And the english subtitles (from .srt) look right. The Japanese subs are from a .sub file.

Any suggestions?

One thought is that maybe the .tsv doesn't match the way that I set up the deck, following nukemarine's tutorial:

nest0r Wrote:A lot of people seem intimidated by subs2srs, especially when they see all the comments in the thread of people trying to get it working on Linux or Macs or whatever, but it's in fact very easy. I recommend just snagging the English and Japanese subs for a single episode of an anime you have, from kitsunekko, and using that to go through the step-by-step instructions cb4960's provided (I only chose anime as a quick trial to get you started, since it's a small file and easy to find). It'll take less than 5 minutes, I swear.*

*Actual time may vary.

I think Nukemarine might have a video up, also? Though I think he use a dorama rather than an anime for an example:

There may be others reasons nestor . I,for one,while very supportive of sub2srs on the beginning , dropped sub2srs about 5 months ago after realizing three things

- language is all about context
what's the point of having all those cards while most of them don't mean s... if they re isolated ???? I have some old schoolrumble card .... I have not the slightest idea of the context . You get a grammatical structure / the numerous meaning of a word because you find it in different context that you really understand , with some level of emotional implication .
Besides use sub2srs on a couple of show and soon enough your database will be crumbling under the load .
those sub2srs cards are all about quantity , not quality .

-language is about massive exposition , which means lot of time devoted to read/listen/watch : sub2srs takes A SHITLOAD of time . A time devoted to sub2srs not to japanese .
I still feel nauseous when I think about the time it took me to create a Mononoke deck . I really do : subtitles off timing , subtitle whose character doesn't appear , video format not supported , etc... Even without technical problem , it takes way too much time . You can also use subtitles you find head first but : 1 it still implies someone created japanese subtitles . 2 8 out 10 times the timing is off . And it's not always something you can correct through a slight global retiming of 2 seconds in aegisub . Subtitles for screen and for sub2srs have a different purpose and timing . So it's either massive retiming or sentence unproprely cut . If you can't even have the audio right what's the point ?

- learning in general is about progression
Real japanese , native japanese isn't naturally designed for SRS . Sometimes it fits with SRS like set expression . most of the time it doesn't .
You have to break a sentence to take what you can grasp . And by break I don't mean conveniently cut it in half , I mean really rip the sentence apart : strip it off its adjectives , change subject to make it clear or create an association , etc...
You can't do that with audio . It's set in stone


you're better off watching your favorite anime/drama/movie again and again and again . Then ripping of the audio on your ipod . And creating adapted text cards for specific part
I m still very grateful of cb4960 effort . It's good creative thinking . And it made me aware of some language features . But I can't agree with the concept anymore .

foodcubes Wrote:One thought is that maybe the .tsv doesn't match the way that I set up the deck, following nukemarine's tutorial:

Sounds like you're not mapping the fields correctly with the .tsv when importing the deck? Might be easier if you use the template deck cb4960 provides with subs2srs. subs2srs should also produce a pop-up window when it's finished that tells you what goes with what in terms of Audio, Tags, Expressions, etc. I think this is covered on the subs2srs page. This is an Anki thing, though, importing decks, that is, so you'll want to learn that basic aspect of Anki to make your life easier. ;p

@ghinzdra - I think you're wrong in many ways and I also disagree with your conceptions almost completely. You get plenty of context. Context is augmented; this is how the program works. Rote learning is slow, it's not an all or nothing equation. You can integrate subs2srs with your native media intake to boost your learning. Making decks takes only a few minutes. If a source is causing you to take more than a few minutes making a deck, don't use that source.

Anyway, I was speaking of people who want to use subs2srs (the decks made with it and/or the software itself) in the comment you replied to—I was not speaking of the possible reasons why some people sadly do not want to use it in their studies.

nest0r Wrote:

foodcubes Wrote:One thought is that maybe the .tsv doesn't match the way that I set up the deck, following nukemarine's tutorial:

Sounds like you're not mapping the fields correctly with the .tsv when importing the deck? Might be easier if you use the template deck cb4960 provides with subs2srs. subs2srs should also produce a pop-up window when it's finished that tells you what goes with what in terms of Audio, Tags, Expressions, etc. I think this is covered on the subs2srs page. This is an Anki thing, though, importing decks, that is, so you'll want to learn that basic aspect of Anki to make your life easier. ;p

@ghinzdra - I think you're wrong in many ways and I also disagree with your conceptions almost completely. You get plenty of context. Context is augmented; this is how the program works. Rote learning is slow, it's not an all or nothing equation. You can integrate subs2srs with your native media intake to boost your learning. Making decks takes only a few minutes. If a source is causing you to take more than a few minutes making a deck, don't use that source.

Anyway, I was speaking of people who want to use subs2srs (the decks made with it and/or the software itself) in the comment you replied to—I was not speaking of the possible reasons why some people sadly do not want to use it in their studies.

I m cool with your disagreemnt . Actually I enjoy the fact that we can have those kind of discussion without any risk of flame .
just curious about 2 parts

"I also disagree with your conceptions almost completely."
among those 3 propositions with which one you don't agree and why? I' m speaking in a general way , not specifically sub2srs
-language is all about context
-language learning is about exposition (I didn't say all)
-learning in general is about progression

what exactly do you consider "rote learning " ?

I consider watching and rewatching something over and over again to be rote learning. Just watch it once, and if you find enough interesting sentences, spend a handful of minutes running it through subs2srs, and voila, you're done.

What you can do from there is a multitude of things according to your preferences, but the point is that you have highly contextualized information to be deconstructed, discarded, fed into other areas of your Japanese self-study/exposure, etc. It's a highly flexible tool. I've outlined how I use it in other threads, but I didn't feel like repeating them and am too lazy to link.

(Ditto for not wanting to rehash old subs2srs arguments about context, or peoples' tendency to make immersion and SRSing or sentences and words an either/or versus equation, which it isn't.)

Making and using subs2srs decks is the most effective and the most minimal aspect of my regimen.

ghinzdra Wrote:I,for one,while very supportive of sub2srs on the beginning ...

Well alright! Another satisfied customer!

ghinzdra Wrote:dropped sub2srs about 5 months ago ... what's the point of having all those cards while most of them don't mean s... I have not the slightest idea of the context ... your database will be crumbling under the load ... all about quantity , not quality ... sub2srs takes A SHITLOAD of time ... I still feel nauseous ... I really do ... subtitles off timing , subtitle whose character doesn't appear , video format not supported .... it takes way too much time ... If you can't even have the audio right what's the point ? ... native japanese isn't naturally designed for SRS ...I can't agree with the concept ...

Do'h.

it must be hard not to take it personaly . But I REALLY appreciate the work you did with this software . I still learned valuable things and am extremely grateful of what you did . Maybe at some point I'll change my mind again and find a proper way to use your software . It's just that right now I can't .

ghinzdra Wrote:There may be others reasons nestor . I,for one,while very supportive of sub2srs on the beginning, dropped sub2srs about 5 months ago after realizing three things

... time devoted to sub2srs not to japanese. I still feel nauseous when I think about the time it took me to create a Mononoke deck...

I agree with this part. Hence the thread title, "Online Library of subs2srs decks". I spent a big chunk of the weekend hunting for subtitles and videos, and then trying to make a deck. But hopefully I will get good at making decks, so that I can whip them up quickly.

Also hopefully we can get a standardized online subs2srs library going, since it has to be faster to download a ready-made deck than making one yourself.


But I'm really liking the subs2srs decks I've downloaded. I would actually like to make subs2srs my foundation for sentence mining. I'll use a subs2srs deck as normal, and then clip the text from sentences as needed. I can also watch the movie/drama several times, too.

cb4960 Wrote:

ghinzdra Wrote:I,for one,while very supportive of sub2srs on the beginning ...

Well alright! Another satisfied customer!

ghinzdra Wrote:dropped sub2srs about 5 months ago ... what's the point of having all those cards while most of them don't mean s... I have not the slightest idea of the context ... your database will be crumbling under the load ... all about quantity , not quality ... sub2srs takes A SHITLOAD of time ... I still feel nauseous ... I really do ... subtitles off timing , subtitle whose character doesn't appear , video format not supported .... it takes way too much time ... If you can't even have the audio right what's the point ? ... native japanese isn't naturally designed for SRS ...I can't agree with the concept ...

Do'h.

Did you invent subs2srs? I have to say, thank you! It really is an amazing concept and program. I'm looking forward to teaching subs2srs to any and all of my friends who are studying new languages. I also think that you deserve a government grant for developing/to develop the software. Maybe for distribution to public schools.

nest0r Wrote:

foodcubes Wrote:One thought is that maybe the .tsv doesn't match the way that I set up the deck, following nukemarine's tutorial:

Sounds like you're not mapping the fields correctly with the .tsv when importing the deck? Might be easier if you use the template deck cb4960 provides with subs2srs. subs2srs should also produce a pop-up window when it's finished that tells you what goes with what in terms of Audio, Tags, Expressions, etc. I think this is covered on the subs2srs page. This is an Anki thing, though, importing decks, that is, so you'll want to learn that basic aspect of Anki to make your life easier. ;p

Ok, I think I found the problem. For some reason the .tsv file has an additional entry in the Subs2 field. This happens at exactly the third card, which would explain why only two cards are imported. Any suggestions on what I'm doing wrong?

Tag Crows_Zero_1
Sequence 1_003_0.0.40.280
Audio [sound:Crows_Zero_1_0.0.40.030-0.0.42.072.mp3]
Snapshot <img src="Crows_Zero_1_0.0.41.051.jpg">
Subs1 <img src="Crows_Zero_001_Stream_00_Subs1_000.40.28-000.40.91.png">
Subs1 <img src="Crows_Zero_001_Stream_00_Subs1_000.41.01-000.41.82.png">
Subs2 Ken...


Or, if this is too tricky to figure out, could you recommend a positive control. Perhaps a short drama with both Japanese and English subtitles that you know works with subs2srs?

In the .tsv file, Anki separates things like Audio and Snapshot using semicolons or tabs, which is not the case for the two img src thingies you have listed above, right? They're both just the Expression or whatever. So in other words, when you import into Anki and are mapping them, each bit of stuff between the separators will be considered one field.

And by the way, that list subs2srs throws out is just a guideline, like the template. The basic idea is to just make sure the fields in the column match the ones in your deck according to their purpose, and ignore any you don't want to use, if any.

Which again is an Anki thing, so if you're just learning stuff like how/where to procure videos and subs as well as Anki, I don't think it's fair to use your learning a particular set of new things at once in a cpl days as a premise for why we should have an online database of decks. Don't get me wrong, to some degree I agree with the database idea in general, but if the reasons for having it is to avoid people learning how to make their own, I'm against that.

Where were we. I also noticed you used timing from Subs2. Any particular reason why? If both generally match to the video, I might recommend sticking with using the Japanese subs for that, as a general rule of thumb.

nest0r Wrote:In the .tsv file, Anki separates things like Audio and Snapshot using semicolons or tabs, which is not the case for the two img src thingies you have listed above, right? They're both just the Expression or whatever. So in other words, when you import into Anki and are mapping them, each bit of stuff between the separators will be considered one field.

I get it. You're right, they're only separated by a space and not a tab. I just learned that the .idx subtitle format is image based. So it must be a reference to two images.

Yes, vobsubs (the .sub file; the .idx which accompanies it contains the timing info and stuff like that) are images of the subtitles that haven't been OCR'd (converted to text). That makes it annoying for when we want to copy/paste the text but unfortunately that's how Japanese movie subtitles are, in large part. You can always try cb4960's program Capture2Text, but let's not get into that now.

Hopefully by now you've looked at the Japanese subtitles at d-addicts and seen how much stuff there is out there. Those are .srt format, with selectable text.

This stuff is useful to know anyway if you're also going to be watching things with Japanese subtitles, or importing information into Anki based on Japanese you have in .tsv or .txt files (tab/semicolon-separated).

By the way, you've using the Preview in subs2srs, I hope? ;p

@ghinzdra: the lack of context is why cb4960 added the option of adding extra context lines into your subs2srs deck. They can be used with just reading or with audio too. It's really helpful!!!

Right, so, it's a long time since i last did anything with subs, & i've had to reset my computer since then. I know i asked this before, but i have no idea where...

i'm opening the となりのトトロ subs from kitsuneko.net in notepad, but all i'm getting is junk. So far, i've tried:
* resaving the file as a UTF-8 file. (actually, i've tried all the options).
* checking that it's in a japanese font. (i've tried a few).
* changing "current system locale" option in Languages and regional settings to Japan.
* tried opening the subs in Aegisub instead, including with chartype UTF-8, and just get "unknown file type".
* downloaded the subtitles from modelino, but exactly the same thing happened.
* changing the filetype from .sub to .srt, but then i get a parse error.

Am i doing something wrong, or are these subtitles junk?

Hope someone can help!!!

IceCream Wrote:@ghinzdra: the lack of context is why cb4960 added the option of adding extra context lines into your subs2srs deck. They can be used with just reading or with audio too. It's really helpful!!!

Right, so, it's a long time since i last did anything with subs, & i've had to reset my computer since then. I know i asked this before, but i have no idea where...

i'm opening the となりのトトロ subs from kitsuneko.net in notepad, but all i'm getting is junk. So far, i've tried:
* resaving the file as a UTF-8 file. (actually, i've tried all the options).
* checking that it's in a japanese font. (i've tried a few).
* changing "current system locale" option in Languages and regional settings to Japan.
* tried opening the subs in Aegisub instead, including with chartype UTF-8, and just get "unknown file type".
* downloaded the subtitles from modelino, but exactly the same thing happened.
* changing the filetype from .sub to .srt, but then i get a parse error.

Am i doing something wrong, or are these subtitles junk?

Hope someone can help!!!

You might try this from the subs2srs FAQ. It outlines a couple of easy methods.

Edit: Which reminds me - I should add encoding conversion to subs2srs.

thanks!!! ok, i get why it doesn't work now. i gave up already though lol... subrip doesn't recognise it's characters. mm, something easier...

nest0r Wrote:Yes, vobsubs (the .sub file; the .idx which accompanies it contains the timing info and stuff like that) are images of the subtitles that haven't been OCR'd (converted to text). That makes it annoying for when we want to copy/paste the text but unfortunately that's how Japanese movie subtitles are, in large part. You can always try cb4960's program Capture2Text, but let's not get into that now.

Hopefully by now you've looked at the Japanese subtitles at d-addicts and seen how much stuff there is out there. Those are .srt format, with selectable text.

This stuff is useful to know anyway if you're also going to be watching things with Japanese subtitles, or importing information into Anki based on Japanese you have in .tsv or .txt files (tab/semicolon-separated).

By the way, you've using the Preview in subs2srs, I hope? ;p

Yes, I did the preview and it actually looks ok in subs2srs.

I see that IceCream had problems too with .idx/.sub files. Are these inherently tricky to work with? I can try to stick with .srt if that's better for beginners. (Plus, as you said I can't edit .idx text in subs2srs. I'd like to be able to mine from subs2srs/anki by cutting sentences down.)

I tried finding Zettai Kareshi, but there aren't any seeders anymore. So I'll continue the hunt on D-addicts for something else. Do you have any recommendations for a drama with seeders and .srt files?