making head talk

[antoine]

hey everybody,

we modeled a head from photos in 3dmeNow and would like to make it talk in vizard. however were facing a bit of a problem with the preparation of the head to use with the head command in peoplemaker (as described in "Adding Morph targets" and "Making talking head" in worldviz references).

Our question is how we can access the morph (filename: mouth_open) from peoplemaker in the vizard script? Or are we missing out on a step in the proper way it has to be done?

The example lines of code

speech = vizact.speak('jfk.wav')
avatar.addAction(speech)

wont work if we run the script, i.e. the head doesnt talk.

Many thanks in advance!

[Jeff]

Are there any errors showing up in Vizard? Can we take a look at the vzf file? If so, either attach here or use our uploads page and note the file is related to this thread.

[antoine]

hey jeff,

here is the link for the uploading. C:\fakepath\David_5.vzf

thanks a lot

[Kevin Chiu]

Hi Antoine,

Your vzf model was lack of at least one more morph targets.
Because it only hold one mouth open morph, you will need to add one more morph like a normal one and rename the mouth open morph to "mouth_open" to get access with it.
The normal one can be named to "morph" as same as the help file noted.

For your reference.
http://docs.worldviz.com/vizard/p_ad...ph_targets.htm
http://docs.worldviz.com/vizard/p_ma...lking_head.htm

[antoine]

hello again,

we added the mouth-open and morph-targets however we still can't run the lip sync.

here are the links:

C:\fakepath\talk.vzf
C:\fakepath\Therapie.py
C:\fakepath\test.wav

we would be very grateful if you could take a look at our script and check whether there is anything we are doing wrong.

thanks a lot.

[Kevin Chiu]

Hi Antoine,

The model actually works, the problem is that your test.wav didn't have any sound.
SInce the wave file need to have sound to make your avatar's head talking, please change your sound clip and don't forget to call it in your script.

For example,

speak = vizact.speak('jfk.wav') # Use a sound clip from Vizard's resource folder
vizact.onkeydown('a', avatar.addAction, speak)

[antoine]

hey kevin,

thanks a lot.

greetings