OK; Sorry I haven't seen this earlier. I don't always get notifications from here.

In order to keep my sanity using the Rendo search, I've done a quick copy/paste from a thread on making Genesis Visemes for Poser. It's behind a wall. The forum requires a free log in and I forget the rules on linking stuff to here so this missive may be a bit disjointed as I chopped a lot of forum chatter and answers to questions from this grab. I did do a tutorial on how to make and rig a figure from scratch that can be animated to talk. It applies to the rock.

We are going to learn how to make the XML files Poser's Talk Designer needs to do lip sync animation. I've written this stuff before but it's scattered around in various threads. It's really very simple. As I mentioned above, all Talk Designer does is drive facial morphs based on speech phonemes.

Phonemes are the basic sounds of human speech. There are a limited number and it doesn't matter what language you speak. They are all the same. "Baby talk." The noises that an infant makes are the basic phonemes of speech and language is learned by having a parent reinforce the ones that correspond to their language. That means that you can write the xml so the character's lips always sync no matter what the language used

A VisemeMap is the xml file that tells Talk Designer:

The name of the figure it is for.
The internal names of the head and each eye.
Which phoneme morph to drive.
Fine tuning for the head and eyes. and lastly:
Which face files to use for emotional impact. Fear, Anger; and so on.

We will disect Alyson's VisemeMap file to show each of these sections. If you have Poser, you have Alyson's map.

We'll do this in the next post. I've got a plane to catch. pdg

Lots cut out

And just in case everyone thought I'd forgotten. Diminished capacity and all ; I did. Here we go:

~~~ lots of lines deleted ~~~

<Interpolation  spline="1" dropOffTime="0.15" />
<Blink  morph="Blink" openValue="0.0" closeValue="1.0" closingTime="0.167" closedTime="0.034" openingTime="0.134" />
<HeadBendParm  parmName="Bend" maxValue="-0.5" minValue="1.0" energyThreshold="7" minimumInterval="0.12"/>
<HeadSideParm  parmName="Side-Side" maxValue="-0.3" minValue="0.3" interval="0.07" />
<EyeSideParm  parmName="Side-Side" maxValue="5.0" minValue="-5.0" forerun="0.1" minimumInterval="0.8" toSideTime="0.1" staySideTime="0.7" backToNeutralTime="0.1"/>
<EyeUpDownParm  parmName="Up-Down" maxValue="5.0" minValue="-5.0" forerun="0.01" />
<EyebrowParm  morph="BrowUp" maxValue="0.3" minValue="0.0" />

<Emotion name="Anger" file=":Runtime:LipSync:Face:Alyson:Alyson Anger.fcz"/>
<Emotion name="Disgust" file=":Runtime:LipSync:Face:Alyson:Alyson Disgust.fcz"/>
<Emotion name="Fear" file=":Runtime:LipSync:Face:Alyson:Alyson Fear.fcz"/>
<Emotion name="Joy" file=":Runtime:LipSync:Face:Alyson:Alyson Joy.fcz"/>
<Emotion name="Sadness" file=":Runtime:LipSync:Face:Alyson:Alyson Sad.fcz"/>
<Emotion name="Surprise" file=":Runtime:LipSync:Face:Alyson:Alyson Surprise.fcz"/>

This is the Alyson viseme map. I've chopped a lot of the lines from it for brevity sake. We'll take each section and explain what it is and does. By the time we are done you will know how to make any figure talk in Poser. Doesn't matter what it is or which company or person made it. If the figure works in Poser then we can make it talk. Over the years I've done the "Heavies" series of animals, Mil Dragon 1 and 2, A car for a cartoon take off of an old American tv show, My Mother the Car; and a donkey to do a Francis The Talking Mule take off; and the Poser horse of course.

Each XML file has 5 parts as I have mentioned above. They are pretty self explainatory but let's look at each section one at a time.

This is the first section. It tells you, the user; the name of the character the file is for. If we were making this file for UGH by D3D then we would replace figureName="Alyson" with figureName="Ugh".

If we were doing this file so that Ugh spoke German, we would change language="English" to language="German". This is for the user's benefit. Poser doesn't care what the language is.

subLanguage="US" This is the dictionary Poser uses to match phonetics when using typing the text that matches your audio file. The phonetic dictionaries in Poser are editable and you can have as many as you like. They just need to be the one pointed to by the xml file. That allows you to have your characters lips sync with special sounds that are not usually found in English.

This is the second section. It assures that the figure you are animating in Talk Designer actually moves. Because different figures may have different internal names for the head and each eye; this line exists to ensure compatibility with the figure you use. You can check the internal name (the name Poser really uses) of the head and each eye by clicking on them and accessing the properties tab. These internal names are usually the names assigned to the obj file when they were created or grouped.

~~~ lots of lines deleted ~~~

These lines are the majority of the file and the ones that drive the facial expressions. If we have a set of phoneme morphs for our figure this is an easy section to work with. Each phoneme will have a corresponding morph and we just make sure that the internal name of the morph corresponds with the correct phoneme. In the lines shown above the only thing that would change with different figures is morph="?" where the ? is replaced by the internal name of the morph you want to activate a given phoneme. The phonemes never change. They are a fixed part of speech. Only the morphs that a phoneme drives will ever change.

If you do not have a set of phoneme morphs one need not fear. These morphs are the mouth and tongue shapes made when you talk. You can make your own morphs and link them to the phoneme of choice. If you do not have the phoneme morphs or the skill to make your own do not fear. Just hook the phonemes up to open/close mouth morphs and maybe a few smile morphs as well. If you do not have a morph to drive a phoneme just delete the entire line from the file. There is a minimum of one phoneme definition per file but I would recommend you leave in a couple to drive what ever morphs you may have. The more phonemes, the more life like the speech.

This fourth section is a fine tuning section. Talk Designer will add in eye blinks and random head motions to make your speech more life like. If you find that the character you are using is moving too far to the extreams, or maybe you want them to move a bit more; then this is where the fine tuning of the head and eyes takes place. Modify message

The final section. The emotions. These are face files that you link Talk Designer to to fine tune your figures spoken words.fcz is just a compressed fc2. A face file. You can make your own facial pose and save it to the Face Room. You can buy a set of facial emotions. Some may have come with your figure. Just link these emotions to face files of the proper name such as Joy.fcz or Joy.fc2. Spelling matters.

Save the file with a name you can find, and you are done. You have made a Talk Designer xml file that will allow the figure of your choice to be 100% compatabile with Poser's Talk Designer. No matter who made the figure. If you cam move it in Poser then it can talk in Poser. Even Genesis.

Hope this helps.

Forum: Community Center

Subject: LIPSYNC FOR DON AND JUDY