AI Agent Skills

Keep a large headline fixed while the presenter leaves the frame, creating a strong transition from a speaker to the message.

Build a reproducible local rough-cut workflow for talking-head or narrated screen-recording videos with transcript review gates.

Official implementations for paper: DreamTalk: When Expressive Talking Head Generation Meets Diffusion Probabilistic Models

Overlay doctrine for the embedded-captions workflow — the caption MODEL (drop / rail / embed) and the rule that captions are an OVERLAY composited on top of the film, ne…

Create a consent-aware digital-avatar layer for AI video production with narration approval and final quality gates.