Bouncing Karaoke Plugin Available?

[Teleri]

Heya folks,

Lately I've seen a few anime's where they had the karaoke bouncing to the syllables. They seemed to match the swooping effect perfectly so I figured that there might be a plugin for this somewhere. I tried a search on the net and on this forum but I came up empty handed.

Do any of you guys know of such a plugin? Or does such an animal not exist?

Thanks for any replies, my current project is kinda on hold until I get this figured out.

[Yusaku]

Advanced SSA. But you have to do it manually in a few subtitles with same text and different effect (at least that's the way I'm doing it...)

Get it at http://vobsub.edensrising.com

[Teleri]

I've stared at the Advanced SubStation alpha codes for a long time but I havent been able to come up with a way of managing this. I asked a friend of mine whom belongs to a fansub group and he wasnt able to figure it out either.

I'm only familiar with \K \k \kf and \ko in SSA. Is there a switch that carries this out in a fashion similar to a normal \K switch or is it more makeshift than that?

Because the bouncing would have to be based off of the karaoke syllables I always figured that there would be a command for it, but it seems unclear. *Boggle* Ah man, I'm getting a headache. I take about 4 advil a day, not good.

At any rate, any clarifications would be appreciated, I've tried playing with the codes and previewing them in virtualdub with Textsub but I havent been able to get anything remotely close to what I'm looking for here.

I've attached the ASS codes that I know of so far. These are the ones that come with VOBSub as per the url above.

[Yusaku]

say you want to highlight a song consisting of three parts, "kakeki" each lasting for 100ms. Normal text size is 20. You want to move just Y axis (so that the rest of text doesn't move). And you don't want any highlighting (you can add that using normal \k tags).


{\t(0,50,1,\fscy150)\t(50,100,1,\fscy100)}ka{\r\t(100,150,1,\fscy150)\t(150,200,1,\fscy100)}ke{\r\t(200,250,1,\fscy150)\t(250,300,1,\fscy100)}ki


See the pattern? (i just hope it works, havent checked it :p )

[Teleri]

Hiya Yusaku,

Thanks for the code. This is pretty sweet. There is only one thing that I have a question with now though. How would you find the times for each syllable? The Karaoke values (\K67, ect) dont translate into any viewable times as far as i know. The closest thing that I could think of would be giving each syllable it's own event,,, but that sounds too stupid to be practical... and I dont think it would work anyways.

From what I read in the help file earlier the timing values for Karaoke differ from the standard time values which are used to give the bouncing effect.

Your a Wealth of Information, thanks for taking the time to explain this. After I find out how to time this I should be golden. I'm sure that I'm not the only person that wanted to know this. ^_^ Domo!

[Zarxrax]

The value is milliseconds. So 100 is equal to 1 second (am I wrong?). Also, a value of 4 is equal to 1 frame of video (at 24fps). What I do is fill in some placeholder value for each syllable, like 32, and then just adjust by multiples of 4 until its timed correctly. Takes maybe about an hour for a standard OP or ED.

[Teleri]

Gotcha,

That makes sense and sounds workable, but what method do you use of timing each syllable? Do you create an event for each syllable when timing from a WAV file in Substation Alpha?

[Yusaku]

Zarxrax: According to SI table, "milli" means the original unit multiplied by 10E-3, i.e. divided by 1000. So 1000 is equal to 1 second. And 40 is equal to one frame in 25fps PAL stream (which you find almost nowhere), not one 24fps frame. But you can use subresync to change between time and frame based formats (but this discussion is on way more advanced topics, at least I hope so). Anyway, just kidding - do not start flaming, please :)

The easiest way IMHO how to TIME opening fast is to do it for each single syllable at the timing phase; and then just recount it for the events. Timing is unfortunatelly the least time consuming process when dealing with ASS, because the code becomes quite unreadable when you start playing around (see my *simple* example), so timing is not the issue with OPs - designing it will take waaay more time.

[Teleri]

So basically I just time each syllable as a separate event and the K value assigned to it will be the end result since there is only one item to time?

[Yusaku]

nope. You time it as multiple events (because unfortunately that's the only way how to do it in SSA, AFAIK)

Then, you go to the first event, extend it to whole time and start substracting in your head to write the \k tags in the first event.

And when you're done with that (that's fast one...), you start playing hours and hours with ASS tags :)