View Full Version : realtime 3D Avatar

03-09-2011, 09:50 AM

Does anyone know of any software that automatically generates a lip synced 3D Avatar in response to a given speech audio file?


03-09-2011, 10:01 AM
I used something like that many years ago, like in 2000 I think.
It was used to generate a talking, real-time webhost, but limited to a head at the time.
You could feed it a audiofile and it would 'lipsync' to it, although a bit crudly.
If you still need it tonight I will go look through my old works to see if I can still find some reference to the company.

Here is a link to some tools and companies for avatar generators

03-09-2011, 10:04 AM

Does anyone know of any software that automatically generates a lip synced 3D Avatar in response to a given speech audio file?


xbox 360 will have this in april with a kinect;)

03-09-2011, 02:10 PM
Hi Sarford

If you could have a peek around for it that would be great!

Thanks in advance.

03-09-2011, 09:40 PM
CodeBaby might help you but the license is quite pricey. Perhaps check out Reallusion's Crazy Talk and/or iClone

03-10-2011, 01:29 AM
Thanks Grace, I'll check those out.

I was also thinking that Adobe director might be appropriate or is that barking up the wrong tree, any thoughts on that?

03-10-2011, 06:52 AM
Poser can sort of do something like that. Would depend on how accurate you want it to be, I suppose. But it isn't realtime. You "render" the animation providing an audio file and optionally, a script to make it match better.

03-10-2011, 08:31 AM

I think Adobe Director is more geared towards developing multimedia content as opposed to the lip-synced Avatar. As it happens I am also keenly interested in the same thing you are. We are developing eLearning courses at work and I would love to have a lip synced Avatar. I have bought Crazy Talk and iClone and are going to start learning them. Manholoz brings up a good point about Poser. I just ordered PoserPro2010 (yes, I have gone on a shopping spree) and we'll see how that does. I bought the Pro version because apparently it integrates with LightWave so that'll be fun to try out. So much to learn!! But it's all exciting - surly ONE of these solutions will work (I hope) :)

03-10-2011, 09:44 AM
Thanks Manholoz.

Good luck with your learning Grace, I briefly played with Mimic Pro and LW a few years back, it's great fun.

The reason for my search is to find a web based solution that can be streamed at near runtime. This means pre-rendering isn't an option.

I'm currently using Softimage at work which has improved lip syncing and Face Robot. Neither of which I've used yet. After rigging a head up, I feel the whole sherbang needs to then be handed over to some other UI which can combine audio files with given meshes and output a file suitable for web streaming.

It's the other UI and how it's made and with what software is what's stumping me. Maybe Unity, Director, Flash. I guess there'd be a fair bit of additional coding, but it's getting the overall workflow first. :(

03-11-2011, 06:09 AM
Sorry I haven't got back to you Blondimage, been pretty busy.
This weekend I'll probably have some time to look into it.