Spoken dialogue modeling poses challenges beyond text-based language modeling, requiring real-time interaction, turn-taking, and backchanneling. While most Spoken Dialogue Models (SDMs) operate in half-duplex mode-processing one turn at a time - emerging full-duplex SDMs can listen and speak simultaneously, enabling more natural conversations. However, current evaluations remain limited, focusing mainly on turn-based metrics or coarse corpus-level analyses. To address this, we introduce Full-Duplex-Bench, a benchmark that systematically evaluates key interactive behaviors: pause handling, backchanneling, turn-taking, and interruption management. Our framework uses automatic metrics for consistent, reproducible assessment and provides a fair, fast evaluation setup. By releasing our benchmark and code, we aim to advance spoken dialogue modeling and foster the development of more natural and engaging SDMs.
@article{arxiv.2503.04721,
title = {Full-Duplex-Bench: A Benchmark to Evaluate Full-duplex Spoken Dialogue Models on Turn-taking Capabilities},
author = {Guan-Ting Lin and Jiachen Lian and Tingle Li and Qirui Wang and Gopala Anumanchipalli and Alexander H. Liu and Hung-yi Lee},
journal= {arXiv preprint arXiv:2503.04721},
year = {2025}
}