makehumancommunity.org

Hello. I have made a full viseme set, in Target form, for MakeHuman and Blender. I have also added a matching phoneme breakdown set to Papagayo-NG.
Until the merge with Papagayo-NG is complete, you can download the additions from the Hunanbean-Collective Github page https://github.com/Hunanbean-Collective/.
there you will find the MakeHuman targets, and the files you need to copy to Papagayo-NG to allow for a full phoneme breakdown.
I use the MakeShapes addon https://github.com/AssetDK/makeshapes to import the Targets into Blender
I use the LipSync Importer & Blinker https://github.com/iCEE-HAM/io_import_lipSync-blender2.8 to bring these to life in Blender.

I will be creating and posting a series of instructional videos and texts shortly, and will post a link here when they are ready for viewing.

Hi

I just did a quick look, looks interesting

Be careful, the code seems to be Windows only ...

Code: Select all

filename, extension = os.path.splitext(dirName + "\\" + file)

you should change that to:

Code: Select all

filename, extension = os.path.splitext(os.path.join(dirName, file))

to make the linux and MacOS users happy, They use the slash and not the backslash and python os.path is simply encapsulating this difference.

Greetings from another guy on github who used most of his free time to find errors in the last weeks

Thank you for taking a look

my question is, which code are you refering to? the Targets and the Phoneme set on hunanbean-collective github should all be, essentially, code agnostic.
Are you refering to the folder structure i used on github, or perhaps one of the other code bases i linked to?

definitely want to fix any problems a.s.a.p

Thank you

Here is a link to a quick video of a stylized female character head-shot using these visemes

https://youtu.be/Y6pfMW6QyIE

I consider this beta 0.4
I intend to update rather quickly

No more jumping through hoops to get the visemes fully functional on your character!

I have incorporated the full viseme set into the MHX2 importer. Equipping the visemes on your character is now just a checkbox away. (well, that and replacing a file in the plugins folder)

The replacement file is located in the faceshapes folder on my git https://github.com/Hunanbean-Collective/MakeHuman-Additions Along with some basic instructions on installation and usage.

Please enjoy!

I know this is old link, but did you make a tutorial on how to use these visemes? You mention a check box in MHX2, but where? I'm using 1.2.0 MH and Blender 2.92.0 on MacOS.

Just saw your message. apologies for the delay in response.
I did not make a video yet on how to install due a perceived lack of interest, but there are text instructions on the git. They will be a bit different for mac, i would imagine.
Basically, all you need to do is replace the faceshapes.mxa in the blender plugins>mhx plugin folder with the one from my git. https://github.com/Hunanbean-Collective/MakeHuman-Additions
As far as the checkmark, i just meant that all you have to do after replacing that file is checkmark the 'faceshapes' box when importing your mhx into blender. The actual lipsync mapping part is handled by Papagayo-NG which has integrated my expanded viseme set as default, and just today released an automated version with speech recognition! After you export your lipsync data from Papagayo-NG you the use the lipsync plugin for blender https://github.com/iCEE-HAM/io_import_lipSync-blender2.8 and select the mesh, set the 'import file' to the .dat file and set your ease in and ease out to 4. Then you can watch the magic

let me know how it goes, and if you have any issues

Be well!

summary:
Replace faceshapes.mxa file with mine and select faceshapes checkbox when importing to blender (MHX)
Use Papagayo-NG to create the lipsync timing data (As of today, now automated with speech recognition!!!)
Use the Lipsync plugin in blender to read and apply the .dat lipsync data from Papagayo-NG to your MakeHuman made human