AI Paper Atlas

Research map / Video / Video generation

StreamingTalker: Audio-driven 3D Facial Animation with Autoregressive Diffusion Model

Yifan Yang, Zhi Cen, Sida Peng, Xiangwei Chen, Yifu Deng, Xinyu Zhu, Fan Jia, Xiaowei Zhou, Hujun Bao

AAAI 2026 · Video › Video generation · approach: Talking face generation · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add diffusion head instead of direct MLP prediction (ablation)StreamingTalker · BIWIFDD: 5.3934 -> 3.6690Improved
add VQ-VAE latent space instead of raw motion denoising (ablation)StreamingTalker · BIWILVE: 8.4578 -> 4.2504Improved
use VQ-VAE instead of standard VAE (ablation)StreamingTalker · BIWILVE: 4.2686 -> 4.2504No clear effect
use fixed-length history instead of all history (ablation)StreamingTalker · BIWILVE: 4.7598 -> 4.2504Improved
keep cross-attention between audio and motion (ablation)StreamingTalker · BIWILVE: 11.7510 -> 4.2504Improved
keep self-attention layers (ablation)StreamingTalker · BIWIMOD: 15.5354 -> 8.5208Improved
keep self-attention layers (ablation)StreamingTalker · BIWIFDD: 4.1083 -> 3.6690Improved
autoregressive diffusion with fixed history instead of full-sequence diffusionStreamingTalker vs DiffSpeaker/FaceDiffuser · BIWI-Test-B long sequences (2000 frames)LVE: 5.2213 -> 4.4596 (vs DiffSpeaker)Improved
autoregressive streaming generationStreamingTalker · Inference latency, 3-27 s audiolatency: 25 ms first frame, constant across audio lengthImproved
StreamingTalker vs DiffSpeaker on short sequencesStreamingTalker · BIWIMOD: 8.4091 -> 8.5208Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy