Thank you!! This is so awesome!

I don't know if I'll ever be able to go back to my boring old text-only deck now.

I don't want to be enerving with feature suggestions - but for the program to be perfect for me, I would love the following option:
Create a "fine-tuned" version of your deck. What I have in mind is the following: You would be able to preview the sound from the mp3-file while seeing the subs. There could be 4 buttoms/fields where you can enter data:
-delete card (bc. there is some really loud background stuff going on, massive mumbling...)
-adjust start/end time (timing slightly off, using the prog for music videos with selfcreated srt-file of lyrics...)
-keep card as it is.
Going through a file like this would probably take an hour or so - but when you consider, you will spend a huge amount of time on it once it is in anki, it should be well worth it (it definitly would be for me).

Hello,

I have just released version 6 of subs2srs (see the first post).

New Feature:

You can now tell subs2srs to automatically rip and convert the audio track directly from the video file. This should make the program far easier to use because you no longer have to rip the audio yourself (although the option still exists).

It has been successfully tested with AAC, VORBIS, and MP3. I couldn't get it to work when the audiotrack is of AC3 format, but maybe the video file I tested had a DRM protected AC3 file.

Thanks to Killersquierl for pointing me to the utility that makes this possible: ffmpeg.

cb4960

HerrPetersen Wrote:I don't want to be enerving with feature suggestions - but for the program to be perfect for me, I would love the following option:
Create a "fine-tuned" version of your deck. What I have in mind is the following: You would be able to preview the sound from the mp3-file while seeing the subs. There could be 4 buttoms/fields where you can enter data:
-delete card (bc. there is some really loud background stuff going on, massive mumbling...)
-adjust start/end time (timing slightly off, using the prog for music videos with selfcreated srt-file of lyrics...)
-keep card as it is.
Going through a file like this would probably take an hour or so - but when you consider, you will spend a huge amount of time on it once it is in anki, it should be well worth it (it definitly would be for me).

All suggestions are appreciated. I'll see what I can do.

For anyone interested, here is the latest source code:

Download subs2srs version 6 source code on MediaFire

Thanks for the automatic audio update, that makes it easier.

Is it possible to have the program generate readings like in the Toki wo Kakeru Shoujo deck?

Paludis Wrote:Thanks for the automatic audio update, that makes it easier.

Is it possible to have the program generate readings like in the Toki wo Kakeru Shoujo deck?

It shouldn't be a problem to make that an optional feature. I'll see what I can do.

this is clearly the greatest thing since sliced bread, and I will gladly elect you the mayor of srs for your outstanding contributions to the community.

upon trying this out, i went into the dark hole that is gathering subtitles - and it was terrible - japanese subtitles are hard to make and really hard to find

-would it be possible in future versions to use other subtitle formats like .sub and display them in the srs as pictures instead of text? If I could just take the english and japanese .sub files off my dvds, then i could make any number of decks quickly.

thanks for everything that you have done up to this point and keep up the good work

duder Wrote:this is clearly the greatest thing since sliced bread, and I will gladly elect you the mayor of srs for your outstanding contributions to the community.

upon trying this out, i went into the dark hole that is gathering subtitles - and it was terrible - japanese subtitles are hard to make and really hard to find

-would it be possible in future versions to use other subtitle formats like .sub and display them in the srs as pictures instead of text? If I could just take the english and japanese .sub files off my dvds, then i could make any number of decks quickly.

thanks for everything that you have done up to this point and keep up the good work

Good suggestion. Might be a while before I get around to it though.

I can't work out how to import into Anki. Is there any guide for importing tags & making models in Anki because I can't seem to figure it out at all. Could somebody post some screen-shots of Anki for importing subs2srs or point me to some help pages on Anki models? I've looked at the Anki website but the terminology they use doesn't bear relation to the actual program.

this line:
>Then, when importing, map each part of the import file to the appropriate field.

What is a field and what are tags etc :S

BTW please keep the subs2srs program so I can use it for all languages not Japanese-only.

duder Wrote:upon trying this out, i went into the dark hole that is gathering subtitles - and it was terrible - japanese subtitles are hard to make and really hard to find

-would it be possible in future versions to use other subtitle formats like .sub and display them in the srs as pictures instead of text? If I could just take the english and japanese .sub files off my dvds, then i could make any number of decks quickly.

I've been down that dark hole too. Not pleasant.

I second this suggestion. It's a shame that you won't be able to edit the text (since it's an image) but it's better than going down that hole.

Great to see the positive reaction.

In the subtitles downloader program SubDownloader-2.0.9.3.exe you can choose the language - is it not useful for finding Japanese subs?

I've heard the website Shooter.cn is great for Chinese subs - is there an equivalent website for Japanese subs?

So is there a way, in some program or another, to take multiple subtitle files (from different sources) and eliminate lines that contain repetitions of items selected by the user? Kanji compounds, for example, perhaps in lines of a specified length. I'm a database/regex newbie, so.

Or even to create a master file, based possibly on an exported Anki deck, and use that somehow to create an exclusion list for other decks or files to reference. Maybe that sort of thing would be best left to some kind of import plugin for Anki? I'm just brainstorming, here. I guess this last bit goes back to what we were talking about earlier. I'm basically looking for stuff I can do right now to spare cb4960 the trouble of having to fiddle with subs2srs.

anon6969 Wrote:What is a field and what are tags etc :S

Examples of fields:
expression, meanings, reading, audio Clip, snapshot, etc.

To create a field in Anki:
Settings->Deck Properties...->Edit->Add

The Anki import dialog will let you map the Anki field to the location in the import file.

On the VOBSUB front, I haven't found any command line tools to convert to VOBSUB files into bitmaps. However, I can probably rip the relevant code out of SubtitleCreator to accomplish this.

Edit: vsconv+DVDSupDecode might be what I'm looking for.

anon6969 Wrote:What is a field and what are tags etc :S

http://ichi2.net/anki/wiki/KeyTermsAndConcepts

Oh, my mistake, I didn't realize include/exclude was already in the program. You're evolving too fast for me! ;p This gives me some options to tinker with.

In 2007, AJATT recommended splitting audio from TV shows and movies into 3:30 minute segments and putting them on your iPod (allowing for random play, or easy to start at middle segments). I have had pretty good experience do this for dramas I liked.

It occurred to me recently that this updated program is perfect for stripping audio for the iPod. Guess it's a matter of having a "subtitle" file that's just evenly spaced times.

Much, much easier than my old: Strip audio with Cool Edit Pro, split audio with Audacity. Plus, this can do it in bulk.

CB4960, if the desire is large enough, think you can add an option where we can just strip audio from the videos in pre-specified lengths like 5 minutes, or 3:30 minutes?

Nukemarine Wrote:CB4960, if the desire is large enough, think you can add an option where we can just strip audio from the videos in pre-specified lengths like 5 minutes, or 3:30 minutes?

Sure. I can probably stick this feature in a tools menu or something.

Just wanted to say I finally used this to make a deck, it works like a charm, thanks again!

Omnistegan Wrote:

stoked Wrote:I'd love to see a linux port of this. Cool stuff.

ooo, yes, I forgot to say that. A linux port would be amazing.

it uses .net so you could try it with mono

This program is awesome.

Will try to see how SRS'ing an entire movie goes.

What would happen if I took untimed Japanese closed captions and set subs2srs to use the English subtitles for timing? Would it magically work, or would I be left with a jumble of mismatched text and audio?

nest0r Wrote:What would happen if I took untimed Japanese closed captions and set subs2srs to use the English subtitles for timing? Would it magically work, or would I be left with a jumble of mismatched text and audio?

Currently, the only valid subtitle inputs are .srt and .ass/.ssa files. So some sort of timing information is required. If the timings between the 2 subtitle files don't match, subs2srs will try its best to get the closest match. But if the timings are not present (or all mapped to the same time), subs2srs will get consfused. With some effort on your part, you could map the closed captions to the English subtitle timings in a program like Aegisub. I don't think that this could be done automatically, since if the two files don't have the same number of lines of dialog, the timings would be way off.

KREVA Wrote:So how would one go about ripping subtitles from a movie into a subtitle file that can be put to good use with this program? If OCR is the only way, I think I'll pass (had bad experiences with using OCR back in the day).

This would depend on your level of Japanese I guess. If I watch a film I tend to understand more sentences than I don't (likely the same with Tobberoth) so removing all could take a lot of time too (Im not even sure how easy it is to remove stuff from Anki in bulk, never tried).

Just wanted to say that this program does look pretty awesome, well done. I'm not on Windows so couldn't try it out, sadly.

Of interest, what did you write this in? Are you planning on releasing the source? (apologies if you already answered that, I skimmed the thread). I'd probably be quite happy to help porting this to Mac/Linux if you wanted a hand.

Cheers