cb4960: Thank you so much for the amazing software!

I don't know if someone proposed it already, but what this software really needs is the ability to merge subtitles. Most of the time when I add cards from anime into the SRS, I want multiple lines to be included in the same card, and right now to get around it is quite the hassle. I got around it by editing the subtitles themselves, just joining them with an editor, but native support, possibly in the preview screen would be a huge advantage.
Thanks for sharing the software!

Absolutely amazing software!

One thing though - on my not all that ancient PC processing even 5mins/80 cards worth of video with sub2rs with a window size of 352*202 pix takes good half an hour. Now I'm not a great programmer, but I wrote a decent number of image processing applications in the past and don't quite see why would it be THAT time consuming, most of software that only slightly alters video is able to run things quicker than it takes to play the video itself. Should I really expect the process to take that long, or is it perhaps something to do with settings that I can adjust?

My machine has a 2ghz processor, meagre 64mb RAM, and runs Vista. I'm asking sub2srs to create a deck spanning 5 min, some 80 cards, with videos in 352*202 pix resolution. I supply it with a video in avi format, resolution of some 750x540 pix (roughly, I'm at work now, so can't check it exactly) and subtitles in text format.

Any ideas here? I can live with such a long processing time, it's not like I need to be in front of the PC doing nothing else but staring at processing bar the whole time, but it just doesn't seem right to me. Ah, the bit that takes so long is when a console window comes up and writes out messages about frames, I guess that's the ffmpeg part. Also sub2srs runs in a very reasonable time when asked to work just on audio and snapshots.

Last edited by Puchatek (2011 September 09, 2:23 am)

Overture - worse, I'm slowly falling into dementia and messing up RAM and video card RAM... uhkm... anyway, the machine has 3GB RAM, 64mb dedicated RAM video card.

In the past I wrote on this laptop an application in OpenCV which resized a video to a quarter of its original size and saved that to disc, so essentially same thing as what I'm asking sub2srs to do. It run real time, taking as input image from a camera, and displaying it on the screen in both original and decreased resolution + saving on the disc, all without any visible lag, so I'm really surprised at the amount of time sub2srs is taking to extract videos. Makes me suspect there's some setting somewhere that makes things so slow - something like copying say 50 frames per second (just an impression I got from looking at the console window in action) rather than 20 or so that would be perfectly fine for srs cards. Any ideas here?

Last edited by Puchatek (2011 September 10, 6:29 am)

overture2112 wrote:

Puchatek wrote:

My machine has a 2ghz processor, meagre 64mb RAM...

I assume the 64mb of ram is a typo?

Note, if you have multiple cores and the memory to support it, you can run multiple instances of subs2srs now.

I hope that's a typ, I have 6gb ram, 6mb would be deadly slow

great update. I'd love to use the video feature now that the processing is 2.5 faster. it's just... videos & anki dont work well together for several reasons.

any chance you could include the subtitles into the videos (optional hardsubs/embedded)? first sub in white and the second in yellow beneath, for example.
you could just watch the hardsubbed videos on your tv/pc/ipod without anki and still have exposure to Kanji, not just video/audio. that would be pretty awesome, especially when you're semi- busy doing something else. I hope the hardsubing process is not too complicated.

I also just tried the shift timings feature. its pretty cool but it shifts both Start AND End times. Most subtitles tend to start with some delay so subs2srs cuts out the beginning of the sentence quite often. I used to ignore most vocab that comes at the beginning because it usually doesn't have audio. Being able to shift start times only (-500 default for example) like in aegissub would be very helpful.

thanks again for 24.0!

Any chance we'll get a Mac version in the future or not? ^^

There are 3 programs left I still need to use windows for, subs to srs being one of them. ^^

Great program by the way, until now you have made my life so much easier!

cb4960 wrote:

You can use the Dueling Subtitles feature to create the subtitles and then read this guide to learn how to create hardsubs

I was hoping subs2srs could do that since it processes all the needed video/subtitle info. But I have no idea about programming I just thought it might be easy to include the subs into the videos.
hardsubbing so many videos seperately would take a lot of time.

You can use the Pad Timings feature. To shift the start times only, enter a positive value into the Start field and enter 0 into the End field.

oh, i entered a negative number, thats why it didnt work. ty!

can I save my own default numbers in settings.txt? I opened the file and changed the default audio bitrate successfully but couldn't find [Pad Timings, Snapshot width/height, Output Directory & Name of the deck].

File>Save works but doesn't autoload my save file like settings.txt

keikei wrote:

can I save my own default numbers in settings.txt? I opened the file and changed the default audio bitrate successfully but couldn't find [Pad Timings, Snapshot width/height, Output Directory & Name of the deck].

File>Save works but doesn't autoload my save file like settings.txt

Solution:

1) Create a shortcut to subs2srs.exe.

2) Right-Click the shortcut, select properties, edit the Target field and put the name of the .s2s file to load on startup as the first argument. Example Target field:

"G:\Temp\subs2srs\subs2srs.exe G:\Temp\subs2srs\s2s\startup.s2s"

Last edited by cb4960 (2011 October 20, 1:43 pm)

Wow! I just found your program and I must say it is genius.  It is about to make my life way easier

I was wondering if there is an easy way to merge 2 cards together?  For example when the speaker says a long sentence and the subtitle is cut in half and shown over 2 screens. It would be awesome to splice those 2 audio files back together.  It would also be good because the translation is usually the in the opposite order when it covers 2 screens.

Thanks!

(Sorry if this has been addressed already in the thread... I haven't waded through all of it yet...)

I'm a bit skeptical on the utility of this program. From what I've heard the language used in TV and games is very unnatural and odd-sounding compared to real spoken & written Japanese. Is there much point in learning the language in such environment, at least in the long term?

Betelgeuzah wrote:

I'm a bit skeptical on the utility of this program. From what I've heard the language used in TV and games is very unnatural and odd-sounding compared to real spoken & written Japanese. Is there much point in learning the language in such environment, at least in the long term?

Yes

It's not so much that the language is completely unnatural, as that it is sometimes a bit over-dramatized for effect. It's just like how in English, everyday language is not the same as what is used in something like the Lord of the Rings. As long as you keep this in mind, there should be no problems.

This doesn't mean that you cannot learn from it, either. Language in fiction is still syntactically correct, and written for natives. If they can understand and enjoy it, why shouldn't you?