Reproduce “Full-Duplex Interaction with Speech, Expressions and Gestures” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Siyin Wang, Wenyi Yu, Terumi Chiba, Xianzhao Chen, Xiaohai Tian, Jun Zhang, Lu Lu, Chao Zhang

ELLSA 2 coordinates speech, gesture, and facial expression during speaking and listening. Its generator uses information from incoming conversation and outgoing speech; for a potentially meaningful gesture, it decides whether one fits, then uses a retrieved example to guide realization. On a curated semantic-gesture task, guidance raises ratings from 2.014 to 2.711 on a 1–5 scale, without establishing universal reliability.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.