I've mentioned this before, elsewhere - but being able to create, manipulate and delete visemes without importing an audio track, or text to speech input would allow "cleaner" facial animation. There would not be a need to spend so much time cleaning up the viseme tracks as is the current workflow. This would allow us to enhance the current range of expressions by fine tuning lip, tongue and jaw movement quicker and easier than we can currently do with set frame, puppeteering and auto-lipsync options.

Agreed. Right now I import the audio, then I immediately delete the auto lip sync keys. Then manual lip Sync.

I would like a few extra phoneme preset keys...

Thank you for the feedback. This will be reviewed by the iClone team.

                                                                

Peter
Forum Administrator

www.reallusion.com

Another spammer.

Peter (RL), can we please have some kind of filtering in place that makes it harder to join the forum?

---

Spam Removed.

I'm not really sure what can be done in the way of filtering unless we restrict access to the forum to only registered users of our software.

For now the best defence we can have is the vigilance of our forum members who report the spam as soon as they see it. For those who already do this I thank you. It is very much appreciated.

 

                                                                

Peter
Forum Administrator

www.reallusion.com

I hate to dig up an old thread but is there a solution for this yet?

If not, I have written a C# app that can take English text and convert it to any series of viseme images you can find. See: http://www.speech.cs.cmu.edu/cgi-bin/cmudict - free phoneme dictionary.

It works like this:

1. You load in images of any visems you need (in this case, I broke out the images of the 3Dxchange/iClone visemes using a paint program).
2. You assoicate text to the visemes (ex. EE, Er, Oh, S.Z, etc...)
3. You map 39 phonemes to the visemes.
4. You type in some words. Internally this is converted to phonemes first and then looks up the associated mapped viseme.
5. You can now view/export the viseme string list.

Now you can more easily assign these visemes within iClone.

The reason I posted this is that I'm hoping that this method can be added to iClone...

Can I get a free upgrade when you implement it???

Turns out it's pretty easy to do.

The trick is that you cannot add your own visemes without a "container" for them to be created in. So, make a note of your movie length, and record silence for that length of time. Now, when you open the "lips" track under the "Face" tab in the timeline, you can click anywhere within the blue box and insert a viseme from the pop-up viseme window.

To delete all visemes from the track, drag over all of them. Double-clicking the lips track does not do it in this case.

@steve: the viseme assignment definitely needs improvement, and using text to support that can help.

A long time ago there was a program called LipSync for lip-synching in Poser, which later became Mimic and was bought by DAZ.

You would load a sound file with the dialog as well as the dialog text, and the program would create the lip-synch animation. It worked pretty good.  

I have suggested something like that as a feature request quite a while ago, so it's good for you to bring this up again. As it is now, the viseme assignment in iClone is about 90% inaccurate. The mouth moves more or less in time, but with the wrong visemes, so you need to do a lot of cleaning up to get it right.

BTW, I don't work for RL, so I can't promise you a free upgrade...

---

Agreed.

It would be nice to be able to add visemes directly to the "Lip" sub-track without requiring a "Voice" segment.

---

On a related note, I entered a request that the viseme creation algorithm be "calmed down" a bit.  I think it makes too many, and the lips look like they're being electrocuted when the avatars talk.  (Okay, not all the time, but a lot of the time.)

https://forum.reallusion.com/Topic190151-298-1.aspx

---

iClone 7... Character Creator... Substance Designer/Painter... Blender... Audacity...
Desktop (homebuilt) - Windows 10, Ryzen 9 3900x CPU, GTX 1080 GPU (8GB), 32GB RAM, Asus X570 Pro motherboard, 2TB SSD, terabytes of disk space, dual  monitors.
Laptop - Windows 10, MSI GS63VR STEALTH-252, 16GB RAM, GTX 1060 (6GB), 256GB SSD and 1TB HDD

Good suggestions!

And do stay away from TTS if at all possible. I'm pretty sure it does not parse text to viseme directly, but creates the TTS audio and lipsyncs to that.

Those with an older Crazytalk: Did it work better when Crazytalk created the viseme file, and then it was imported into iClone?