Mesqueeb Wrote:Hi!This is where I placed mine:
Since subs2srs doesn't work for me know somehow, I guess I need to do this:
http://miszacus.net/index.php?a=uiaatkbridge
But where do I put the file I've downloaded then? xP
Seems stream 0 codec frame rate differs from container frame rate: 23.98 (65535/2733) -> 23.98 (24000/1001)
Input #0, avi, from '/media/Tracy/動画/中国 普通话/Cape No.7 - tradsubs/cowry-cn7-cd1.avi':
Metadata:
ISFT : VirtualDubMod 1.5.10.1 (build 2439/release)
Duration: 01:03:38.06, start: 0.000000, bitrate: 1531 kb/s
Stream #0.0: Video: mpeg4, yuv420p, 656x272 [PAR 1:1 DAR 41:17], 23.98 tbr, 23.98 tbn, 23.98 tbc
Stream #0.1: Audio: ac3, 48000 Hz, 5.0, s16, 448 kb/s
Unrecognized option 'vf'
** (subs2srs.exe:4437): WARNING **: CreateProcess: error creating process handleMesqueeb Wrote:thanks ! this worked.The "[mpeg4 @ 0x8b83e30]Invalid and inefficient vfw-avi packed B frames detected" message is normal. It's just ffmpeg doing its thing.
But now it gets stuck while Generating snapshots.......
"[mpeg4 @ 0x8b83e30]Invalid and inefficient vfw-avi packed B frames detected"
Is a line I keep noticing
and where the final frame stopped: at generating snapshot 264 / 770
Thanks! ^^Code:
Seems stream 0 codec frame rate differs from container frame rate: 23.98 (65535/2733) -> 23.98 (24000/1001)
Input #0, avi, from '/media/Tracy/動画/中国 普通话/Cape No.7 - tradsubs/cowry-cn7-cd1.avi':
Metadata:
ISFT : VirtualDubMod 1.5.10.1 (build 2439/release)
Duration: 01:03:38.06, start: 0.000000, bitrate: 1531 kb/s
Stream #0.0: Video: mpeg4, yuv420p, 656x272 [PAR 1:1 DAR 41:17], 23.98 tbr, 23.98 tbn, 23.98 tbc
Stream #0.1: Audio: ac3, 48000 Hz, 5.0, s16, 448 kb/s
Unrecognized option 'vf'
** (subs2srs.exe:4437): WARNING **: CreateProcess: error creating process handle
s)" option and put the timestamp from the snapshot into the "Start:" field. In the "End:" field, put a time that is greater than or equal to the end time of your video.
Mesqueeb Wrote:Okay, it says: generating video clip, and it seems to go correctly...Okay, I think I know what the problem is now. In the same folder as subs2srs.exe there is also a file named settings.txt. Edit settings.txt and search for a setting called "force_legacy_ffmpeg". By default it is set to "false". Please set it to "true". This will enable compatibility with the older version of ffmpeg that is in the Ubuntu 10.10 repository. Now just restart subs2srs and and everything should work fine. To be honest, I meant to set it to "true" by default for the Linux release but completely forgot.
BUT! When I look at the terminal it says:
(look at the last line)
FFmpeg version 0.6-4:0.6-2ubuntu6, Copyright © 2000-2010 the FFmpeg developers
built on Oct 5 2010 22:35:47 with gcc 4.4.5
[...]
/tmp/~subs2srs_temp.avi: No such file or directory
Mesqueeb Wrote:Does it have anything to do with the dimensions of the generate video and snapshots? Should we put the dimensions the same as the movie, or keep the standard ones? The crop is also something I don't quite understand... >,<subs2srs won't let you pick invalid dimensions so you should be fine in that regard. I usually pick dimensions that are about 1/3 the video size. The crop will remove pixels from the bottom of video based on the original video resolution. So if your video was 640x480 and you chose a crop of 100 pixels, the final resolution before being resized would be 640x380. Cropping the bottom like this can be useful for removing hard subs that have been burned into the video itself.
Mesqueeb Wrote:Everything works now!I don't know about that one - you might try the Anki forums. Let me know what you find out and I'll put it into the FAQ. If it helps, the video files that subs2srs outputs have XviD encoded video and MP3 encoded audio.
1 more question:
Will the video clips play on the ipod touch's AnkiMobile application? ^^
nest0r Wrote:I just had an idea. For generating video/snapshots, how about an option to use percentages based on the source video file, for quickly determining proportionate dimensions.It's actually not your fault. You are the victim of an ingenious plot of mine. Last July I leapt forward through time anime-style, saw your post today, leapt back, and added the feature. I've been waiting 7 months just for the day when you would be publicly humiliated. I even had to train super intelligent primates to hack Icecream's account, delete her posts, and get her banned - predicting of course that Fabrice would turn off the delete post feature - all so you couldn't delete your post in shame. Muhahahahahha.
Also still hoping to one day see .trs support, though I suppose it's so low on the priority of desires even I forgot to mention it again till now. ;p
http://forum.koohii.com/showthread.php?p...0#pid95340
Edit: Oh right. Snapshot dimensions chooser. Where have I been. I always use the same dimensions, so. Yes. That's my excuse. Plus it's only been in subs2srs since umm July, 2010. I totally knew about it actually. I was just testing you all.
nest0r Wrote:Haha. Very clever... except for one small detail. You will notice that, this forum not being a draconian, irrational system, I could have edited my comment at any time. But I did not. Why? I will tell you! I left my seeming error intact so that it would appear to you that your plan succeeded! I knew all along about your little scheme.Drat! Always one step ahead!
Now you have revealed your devious nature for all to see. Or should I say, I revealed it by laying a most black and cunning complot. I believe there are a few agencies that would be interested in your temporal abilities. Check and mate.
nest0r Wrote:By the way, even though I can't remember why I wanted ‘multiple spans’ as an option, I just wanted to point out that you never incorporated it: http://forum.koohii.com/showthread.php?p...1#pid88561I did implement multiple spans. Kinda. The preview dialog allows you to activate and deactivate lines using the button on the right. You can activate/deactivate one line at a time or multiple lines at a time.
nest0r Wrote:By the way, having never used it, does that extract audio by subtitles thing name the files based on the lines? Or however people use mp3 tags and filenames in playlists (I've never extensively done such a thing) in their iPod Touches and whatnot.The file names contain the name, episode and timestamp. The subtitles are embedded in the lyrics tag.
nest0r Wrote:Ah ha! I figured out what I was talking about before.I suppose you could use also use Aegisub to time the entire conversation as a single 2:37 line and repeat this process for each interesting conversion in the episode. Then when subs2srs does it's normal processing, you would get nicely chopped up audio files in the output directory - each one containing a full conversion.
So it's like, what I was thinking of was to extract the audio and cut it into segments based on spans selected not by time but by subtitle start and end markers—and not with each line between them broken into separate audio files.
So you'd end up with say, a single 2:37 audio file that starts with a subtitle line where someone asks a question, goes through the exchange of dialogue, then ends with the final line in the conversation, just as an example. Or rather, a set of audio files like that but from a single subtitle+video.
Because with current extract audio options, it seems like you'll have clips based on amounts of time rather than meaningful spans of time, and lines would likely be cut off. If we use the audio clips in the deck creation process, you'd have titular/numerical lines that would be consecutively numbered but not necessarily based on consecutive lines from the source, due to filters.
Still, now that I write that out, I think it would probably just be better, rather than developing a whole functionality around a single user strategy, to use subs2srs' current function to generate audio clips from the subtitle file as per the usual deck creation utilities, and then use the preview to make sure a desired string of lines are selected, and simply keep them together in folders or playlists or combine them as a single file with Audacity or something. In that case you could just ignore this comment. ;p
nest0r Wrote:If I was going through that process purely to create audio clips, I'd probably just not use the filters when doing so.You're in luck, that's how it currently works. Just change a filter in either the Main dialog or the Advanced Options dialog and click the Regenerate Preview button in the Preview dialog to see which lines get deactivated.
But as a general idea that occurred to me, based on that preview interface screenshot: Could there be a way to enable/disable the various filters from the preview dialogue to quickly see how they affect lines, or would that be more trouble than it's worth. Or perhaps one could disable the filters for selected lines? By filters I'm referring to Advanced Options like Prune Subs and Language Specific...
rigol Wrote:Maybe a very dumb question, but... I wanted to create a deck for an episode of Air, started it yesterday and it still runs. It's 376 lines, and the snapshots seem to take forever - running for 22 hours now, I let it run overnight and it's at "Generating snapshot: 358 of 376" Why is it taking sooo long? With Clannad, it took about 20 minutes I think. Is it because of some codec?Wow! It might be time to retire that Apple II
. I hope it at least worked OK when it eventually finished.).