Prime Automatic LipSync Tool
Prime AutoLipsync is text to animation Engine olso Audio Animation engine , Auto/manual timing systems and new Voice Recognition System .
how it works:
1st make some morph targets with basic Phonemes A , E , O , L , M ,F , others if you add WQ , U as seperate Phonemes it will be better or u can assign them to O channel
2nd step : open Prime Autolipsync tool and choose your head with morpher modifier then go to setup mapping section ,click update to update channels list then link Phonemes to your morph channels as shown.
3rd step : save your preset so setup mapping step is Required once only for each head.
then go to automatic lipsync section and have fun
Prime automatic lipsync convert text to animation keys so Load your sound file , listen and write your text then gooo , and you will have your animation
there is 4 ways to control timing
1- start time and end -- keys will be between your start and end time
2- start time and interval --it calculate by defined start time and interval value "interval time is predefined space between keys"
3- slidertime and interval -- animation will start on your current slider time and interval value
4- slidertime and Endtime -- animation will start on your current slider time and end on predefined end time "my favorite"
there is morph limit value you should have it the way you like if its value 100 it will give morph keys 100 value for all channels but its custamizable to your case .
there is some other options morphsliders for manual retouch and other self exploring features.
-- "auto phonemes" option automatic figure out phonatics "AO" "EA" "sh" etc .
when checked off every charcter you wrote will have its own key , better leave it alone :)
and can be exported easily to maya , softimage or any other 3d pacage with FBX
For MAYA , Softimage users just import your model with morph/shapes to max as FBX then use the tool then Exported back FBX with animation.
-- Mirror morph targets section "self exploring"
or checkout 1st 30sec here
Prime Autolipsync Understand English characters (A to Z) any foreign characters , numbers who't work .
For other languages you can write any language with English characters , i think All languages has the same phonemes , just small Tiny Differences .
it will work fine with Arabic , French , Spanish , Portuguese , Chinese and others (All Written in English )
I have some Feedback it works fine with many languages this way.
Test it with your mother Tongue and Give us your feedback .
--try to write all words connected together on pronounce as one word and use space mode to make timing faster.
--Prime Autolipsync doesn't recognize silent characters , it will put a key for them , so its not spelling contest, try to write your keys only .
example "what" word at "auto phonemes" mode on will be 2 keys only "O","T" .
another example "dude" "Grape" "made" it doesnt recognize the silent "E" on the end , so better write it "dod" or "dud" doesnt matter if you did assign "U" to "O" same morph.
olso something like "UP" write it "ap" if u did assign "u" to "o" cause its pronounce "a", you get the idea.
--"auto phonemes" algorism is always trying to reduce keys by binding similar keys , if you didn't want that there is another way to work is to write your own keys , and turn "auto phonemes" mode off, that way all what you write is what you get every character will have a key, in this mode if you write "ah" , u will get 2 keys , so write it "a" and so on .
-- if you know what your doing just disable auto phonemes and write your keys directly , it should be the best way to for anyone any langue .
whats new?( History ) :
--new Mirror morph targets
--space = gap mode
when you write spaces on the text it will give you gap between keys , more spaces more gap some people feedback its annoying
so if you turn it off it will ignore any spaces and no gaps between keys .
--bug fix Timing little messed up when you disable "space = gap"
-- annoying slider time isn't moving when creating keyframes anymore.. its faster now.
--bug fix -- when creating keyframes on "slidertime and Endtime" its reversed if your end time lower than your slider time , fixed .
-- extended demo version you can make more keys ,more Text input now..to have more fun.
--Bug "Incompatible types: 0.0, and undefined " Fixed
Thanx to your feedback
fix "h" charcter now gives "a"
fix the error on counting morph channels setup section when there is empty channels .
Fix spaceing Gap mode , works better now.
save last important options on close and open on last save.
Slider channels only showing your choosen lipsync morph list.
Saving presets as dropdown list to switch faster , to save new preset write your preset name and it will be faster as a list contains all your charcters options, to delete a preset select it and right click the list
still old way save-load settings as file availble.
Prime AutoLipsync is not only a text to animation Engine anymore ,its now Audio Animation engine too , Auto timing system.
-v2.01 update usefull to quick setup already rigged/wired charcter heads
Prime lipsync only deals with "Bezier float" morph channel controllers , when some users try it with wired characers already have some sliders by (reaction manger-wire paramters) it doesn't work or give errors.
I made simple fix 2 ways it should pop up when you press animate button
1st question will ask you if you do wanna extract these targets if you press yes it will extract all the mapped targets for you , you need to add them to morpher modifier manually then re-map everything on the script from the setup panel.
if you press no 2nd method will pop up it will ask you if you want to just convert the channels .
Warning -- this will unwire your rig sliders for sure.
if you want to have both your custom rig slides and the script running , you should do it the 1st way and you only need to do this once .
--Voice recognition Engine-- tested on english only not sure about other langue.
--Auto timing new system. -beta phase- need testing and feedback about it.
How to use:
just do the morpher setup as usual , choose a wav sound then run Auto recognition
audio will be processed , Text will appear , new Auto Timing system will try to make the animation fit with the audio as possible.
My tests has some good results and others not so good , will need some adjastments.
maybe u will have to use the old system in fixing animation at some frames as fit.
This's Voice engyne is still in beta phase , your feedbacks is important.
V2 Requirements: prime Autolipsync v2.0 uses Windows Voice recognition System which is integrated to win 8/7 for sure not sure about older xp system.
Tested on Win8 , I guess it will work on Win 7 with no problem
--Tested on Max 2014 -- max 9 and above should work with no problem.
Limitations: all features availble on demo version but you are limited to certain number of keys , text words for testing & evaluation - non commercial use only .