Diarization Error Decomposition Under Pause Annotation Ambiguity

2026-09-09

AI Summary

本文解决了语音识别评估中因停顿标注模糊导致的误差问题，通过提出一种新的误差分解方法，将标准DER分解为停顿相关和核心误差两部分。

Abstract

Speaker diarization evaluation is sensitive to ambiguity in pause annotation, which can inflate diarization error rate (DER) or obscure genuine model errors. We show that morphological closing, which has been used for pause-tolerant diarization evaluation, discards segment-level distinctions. Instead, we propose an exact, overlap-aware decomposition of standard DER into a pause-attributable component, consisting of errors compatible with pause filling, and a residual core component that can serve as a proxy for intrinsic diarization errors. The decomposition leaves DER unchanged, while the pause-attributable and core components vary monotonically with the pause threshold and eventually saturate. Experiments spanning synthetic transformations, annotation mismatch, cross-domain evaluation, and tight-boundary diarization show that the decomposition reveals error sources not apparent from standard DER.

Problem

• Diarization Error Rate
• Pause Annotation
• Ambiguity

Innovation

• Diarization Error Rate (DER)
• Pause Annotation
• Error Decomposition
• Overlap-Aware