I will interactive talkinghead,livestream,animaze,webrtc avatar,live kit,arkit52,rig3d

Are you developing a web glb capable application based on the talkinghead open source project avatar for live streaming, and you want your avatar to feel expressive, responsive, and stable?

I'LL create 3D avatar from scratch base on your reference photos/2D, accurate real-time topology, uv mapping, optimized with complete integrations for Real-Time, Three js projects app & websites, Live streaming, vtubing and vrchat

The model will include the following named shapes for the talkinghead class to function:

Lip-Sync (Oculus/Viseme Standard):

viseme_sil, viseme_PP, viseme_FF, viseme_TH, viseme_DD, viseme_kk, viseme_CH, viseme_SS, viseme_nn, viseme_RR, viseme_aa, viseme_E, viseme_ih, viseme_oh, viseme_ou.

Facial Expressions (ARKit Standard):

At minimum: eyeBlinkLeft, eyeBlinkRight, jawOpen, mouthSmile, mouthFrown, browInnerUp, browDownLeft, browDownRight

(Full 52-shape ARKit set for high-fidelity emotion)

QA Criteria:

Validation: Must pass glTF Validator with zero errors

Logic Test: Blend shapes must work independently without "pulling" unrelated parts of the face (e.g., viseme_PP shouldn't move the ears)

Naming: All blend shapes must match the names above exactly (case-sensitive)