THE ICASSP 2026 HUMDIAL CHALLENGE: BENCHMARKING HUMAN-LIKE SPOKEN DIALOGUE SYSTEMS IN THE LLM ERA
Zhixian Zhao, Shuai Wang, Longshuai Xiao, Hui Bu, Xinsheng Wang, Hexin Liu, Hung-yi Lee, Lei Xie
Abstract
Driven by the rapid advancement of Large Language Models (LLMs), particularly Audio-LLMs and Omni-models, spoken dialogue systems have evolved significantly, progressively narrowing the gap between human-machine and human-human interactions. Achieving truly ``human-like'' communication necessitates a dual capability: emotional intelligence to perceive and resonate with users' emotional states, and robust interaction mechanisms to navigate the dynamic, natural flow of conversation, such as real-time turn-taking. Therefore, we launched the first Human-like Spoken Dialogue Systems Challenge (HumDial) at ICASSP 2026 to benchmark these dual capabilities. Anchored by a sizable dataset derived from authentic human conversations, this initiative establishes a fair evaluation platform across two tracks: (1) Emotional Intelligence, targeting long-term emotion understanding and empathetic generation; and (2) Full-Duplex Interaction, systematically evaluating real-time decision-making under `` listening-while-speaking'' conditions. This paper summarizes the dataset, track configurations, and the final results.
BibTeX
@inproceedings{icassp2026_theicassp2026hum,
title = {THE ICASSP 2026 HUMDIAL CHALLENGE: BENCHMARKING HUMAN-LIKE SPOKEN DIALOGUE SYSTEMS IN THE LLM ERA},
author = {Zhixian Zhao and Shuai Wang and Longshuai Xiao and Hui Bu and Xinsheng Wang and Hexin Liu and Hung-yi Lee and Lei Xie},
booktitle = {ICASSP 2026},
year = {2026}
}