Other 'vocabularies'?

Discuss Papagayo issues here

Moderators: Fahim, Distinct Sun, Víctor Paredes, erey, Belgarath, slowtiger

Post Reply
User avatar
Peteroid
Posts: 162
Joined: Thu Aug 12, 2010 3:57 am
Location: las vegas, NV
Contact:

Other 'vocabularies'?

Post by Peteroid » Wed Aug 18, 2010 12:58 am

Now that I understand how Papagayo and Switch Layers work, I REALLY like them!

I notice the version I downloaded of Papagayo has (by default) a ten phoneme vocabulary (AI,L,FV,MBP,E,rest,O,U,WQ,etc) for the files it produces. Can one get Papagayo to use a different phoneme vocabulary (which I realize would require some sort of 'text-to-phoneme language' translator besides a clear phoneme-to-token relationship)?

The phoneme vocabulary used in Tutorial 5.2 is much longer... for example it uses 'CLOSED' instead of 'rest' and seems to have a phoneme for every letter of the alphabet. Is there some sort of 'text-to-phoneme' translator for this phoneme vocabulary that Papagayo can use?

I'm guessing if so, one 'tricks' Papgayo by putting a new vocabulary file of the same name where the old one is, since I see no menu option to 'use phoneme language' file (of some sort).

How WAS Tutorial 5.2 done? Possibly manually? ;)
[==Peter==]
funksmaname
Posts: 3107
Joined: Tue May 29, 2007 11:31 am
Location: Nelson, New Zealand

Post by funksmaname » Wed Aug 18, 2010 1:49 am

what exactly are you trying to achieve?
Aside from the standard phenomes you can always right click a word and in the text box at the bottom add whatever you want... it takes the automation out of it but then where do you pan to get a 'text to phenome language' translator anyway!?

what you can do though is use a 2nd voice to exchange lipsync for actual movement triggers as all you're doing is exporting a switch file at the end of it... for example, if instead of phenomes you replaced the word breakdown with 'lookleft blink lookright' as long as you have switch layers that relate to these trigger names the dat file will swap them for you on import.

This is even more useful if used with Ramon's 'lipsync with actions' script which means you can insert actions based on the keywords you choose to put into your dat file...

but at the end of the day - what functionality is it that you are actually missing from the 10 phenome breakdowns? how many mouth shapes can one really create? (or actually need to?)
User avatar
Peteroid
Posts: 162
Joined: Thu Aug 12, 2010 3:57 am
Location: las vegas, NV
Contact:

Post by Peteroid » Wed Aug 18, 2010 3:13 am

funksmaname wrote:what exactly are you trying to achieve?
Aside from the standard phenomes you can always right click a word and in the text box at the bottom add whatever you want... it takes the automation out of it but then where do you pan to get a 'text to phenome language' translator anyway!?

what you can do though is use a 2nd voice to exchange lipsync for actual movement triggers as all you're doing is exporting a switch file at the end of it... for example, if instead of phenomes you replaced the word breakdown with 'lookleft blink lookright' as long as you have switch layers that relate to these trigger names the dat file will swap them for you on import.

This is even more useful if used with Ramon's 'lipsync with actions' script which means you can insert actions based on the keywords you choose to put into your dat file...

but at the end of the day - what functionality is it that you are actually missing from the 10 phenome breakdowns? how many mouth shapes can one really create? (or actually need to?)
I'm in learning mode. So it is very likely I won't need anything else but the 10 phoneme vocabulary of Papagayo. But it doesn't hurt to ask!

Ii's just that the Tutorial (5.2) says the 'vista.dat' file it uses came from Papagoya. It didn't... totally different vocabulary and structured slightly differently (Papagoya only details key frames to minimize size, what was used in the tutorial details EVERY frame even if there was no change). This is why when I tried to use a 'vista.dat' file I generated playing with Papagoya it didn't work... wrong phoneme language!

So it is natural for me to ask if maybe Papagoya was capable of producing a 'vista.dat' compatible with Tutorial 5.2. Being new to this, I can't assume no. Especially since Tutorial 5.2 suggests using it (though I now know it means outside of the tutorial).

But based on what you've posted, I'm guessing anything other than the 10 phoneme vocabulary Papagoya outputs, one has to do either manually (or with something other than Papagoya). Which is fine! :)
[==Peter==]
dueyftw
Posts: 1941
Joined: Thu Sep 14, 2006 10:32 am
Location: kingston NY
Contact:

Post by dueyftw » Wed Aug 18, 2010 9:26 pm

I fine that Papagayo is over kill. If you just use Papagayo exactly as the phoneme are placed on each word you will have way to much mouth movement. I concentrate on the mouth opening and closing to the volume and length of each sound (not word).

I use the two frame rule and think of how the mouth and teeth are showed more then just a phoneme that matches the word spoken.

read this:

http://www.keithlango.com/tutorials/old/lipSync.htm

Dale
User avatar
Peteroid
Posts: 162
Joined: Thu Aug 12, 2010 3:57 am
Location: las vegas, NV
Contact:

Post by Peteroid » Wed Aug 18, 2010 10:00 pm

dueyftw wrote:I fine that Papagayo is over kill. If you just use Papagayo exactly as the phoneme are placed on each word you will have way to much mouth movement. I concentrate on the mouth opening and closing to the volume and length of each sound (not word).

I use the two frame rule and think of how the mouth and teeth are showed more then just a phoneme that matches the word spoken.

read this:

http://www.keithlango.com/tutorials/old/lipSync.htm

Dale
Just read it. Cool, and thanks!

This is the line that I think sums up what he is saying, and the rest basically is in support of it:
There are no letters in speech, only sounds, and the shape our faces take to make those sounds.
I never planned to make a phoneme for every letter, I always knew that was both over kill and the wrong approach. But I did use to think that 'more was better' in the lip sync department, when now I understand why typically 'less is best'. Or rather, impression over detail...

I am actually a believer in minimalism in art, so this works for me! And the side bonus it that it means less work (more paying attention, but less action to accomplish it).

There is a LOT to take in when studying animation. I'm doing this as sort of a side hobby with some aspirations to create work that might be either popular or profitable. Bur I basically just want to have fun and use it as means of expression!

By the way, I'm a professional computer game developer. I also use to program in Shockwave/Director, so I'm very familiar the notions of keyframes, layers, and tweening. So I'm having a lot of fun with AS7PRO so far! :)
[==Peter==]
Post Reply