Neither Silence nor Overlap Is Failure: Intent-Conditioned Evaluation of Turn-Taking in Full-Duplex Spoken Dialogue Models

2026-09-23

AI Summary

该研究针对全双工语音对话模型中的轮流发言评估问题，提出了一种基于说话人意图条件的新评估方法TACT，改进了传统的二元窗口规则。

Abstract

Benchmarks for full-duplex spoken dialogue models score turn-taking with binary fixed-window rules that reward immediate response or silence by completeness of the prior turn. We argue that the appropriateness of a response offset, whether delayed silence or anticipatory overlap, is conditional on the speaker's latent intent, identifiable only from that speaker's behavior. We introduce TACT, a benchmark of 9,728 episodes and 73.2 hours from five dyadic corpora; each episode carries dialogue history, a per-speaker memory profile, and an annotator-derived posterior over six intent classes. Scoring replaces binary windows with a strictly proper threshold-weighted continuous ranked probability score whose weights are intent-conditioned timing kernels fitted to human floor-transfer-offset distributions, proving boundedness, consistency, and binary reduction. Across eleven systems the best model reaches 0.47 against a human topline of 0.86, is nearly invariant to speaker profiles, and TACT agrees with human judgments at Spearman 0.81 versus 0.46 for binary metrics.

Problem

• turn-taking
• intent-conditioned
• full-duplex spoken dialogue models

Innovation

• intent-conditioned evaluation
• TACT benchmark
• timing kernels
• continuous ranked probability score
• full-duplex spoken dialogue