Semantic-Aware Interruption Detection in Spoken Dialogue Systems: Benchmark, Metric, and Model

TL;DR

This paper introduces SID-Bench, a new benchmark and metric for semantic-aware interruption detection in spoken dialogue systems, along with a novel LLM-based model that improves responsiveness and robustness, setting a new state-of-the-art.

Contribution

The paper presents the first real-world benchmark, a new evaluation metric, and a semantic-aware detection model for interruptions in SDS, addressing key limitations in the field.

Findings

Significant reduction in Average Penalty Time (APT) by the proposed model.

Outperforms mainstream baselines in responsiveness and robustness.

Establishes a new state-of-the-art for interruption detection in SDS.

Abstract

Achieving natural full-duplex interaction in spoken dialogue systems (SDS) remains a challenge due to the difficulty of accurately detecting user interruptions. Current solutions are polarized between "trigger-happy" VAD-based methods that misinterpret backchannels and robust end-to-end models that exhibit unacceptable response delays. Moreover, the absence of real-world benchmarks and holistic metrics hinders progress in the field. This paper presents a comprehensive frame-work to overcome these limitations. We first introduce SID-Bench, the first benchmark for semantic-aware interruption detection built entirely from real-world human dialogues. To provide a rigorous assessment of the responsiveness-robustness trade-off, we propose the Average Penalty Time (APT) metric, which assigns a temporal cost to both false alarms and late responses. Building on this framework, we design an LLM-based detection model optimized through a novel training paradigm to capture subtle semantic cues of intent. Experimental results show that our model significantly outperforms mainstream baselines, achieving a nearly threefold reduction in APT. By successfully resolving the long-standing tension between speed and stability, our work establishes a new state-of-the-art for intelligent interruption handling in SDS. To facilitate future research, SID-Bench and the associated code are available at: https://github.com/xkx-hub/SID-bench.