发表机构
NTT, Inc.（日本电报电话公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对说话人日志中伪标签过紧化导致漏检的问题，提出三种改进方法，减少漏检并提升日志及下游多说话人ASR性能。
AI 中文摘要
在松散标签（如带有填充边界或填充停顿的语音片段）上训练说话人日志模型，通常会导致模型输出同样松散。为了获得更紧的边界，已有研究提出基于因果和反因果模型平均输出的伪标签方法。然而，由于伪标签是基于估计的，它们可能遭受过紧化问题，这会增加漏检，并可能作为不可恢复的错误传播到下游任务。本文仔细分析了过紧化的原因，并提出了三种应对方法：(i) 移除停顿填充而非填充边界，(ii) 引入预热阶段以减轻因果和反因果预测起始附近的漏检，(iii) 使基于伪标签的协同训练感知用于最终推理的非因果模型。实验结果表明，所提方法减少了由过紧化引起的漏检，并提高了日志准确性和下游多说话人自动语音识别性能。
英文摘要
Training speaker diarization models on loose labels, such as speech segments with padded boundaries or filled pauses, often results in similarly loose model outputs. To obtain tighter boundaries, pseudo-labeling based on the averaged outputs of causal and anticausal models has been proposed. However, since the pseudo-labels are estimation-based, they can suffer from over-tightening, which increases missed detections that can propagate as unrecoverable errors to downstream tasks. This paper carefully analyzes the causes of over-tightening and proposes three approaches to address them: (i) removing pause filling rather than padding, (ii) introducing a burn-in phase to mitigate missed detections near the beginning of causal and anticausal predictions, and (iii) making pseudo-label-based co-training aware of the non-causal model used for final inference. Experimental results show that the proposed method reduces missed detections caused by over-tightening and improves both diarization accuracy and downstream multi-talker ASR performance.
CommentsAccepted to IEEE SLT 2026
