ACM SIGGRAPH MIG, 2026
Constructing Dynamic Conversations with Neural Back-channel Generation and Voice-line Scheduling

This paper turns independently recorded voice lines into a two-speaker conversation, using a neural scheduler and a back-channel predictor to give animators control over turn-taking and listener responses.
Recommended citationPan, Yifang, Chris Landreth, and Karan Singh. (2026). "Constructing Dynamic Conversations with Neural Back-channel Generation and Voice-line Scheduling." The 19th ACM SIGGRAPH Conference on Motion, Interaction, and Games (MIG 2026).