arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.04078eess.AScs.SD
收藏

基于语音活动和重叠检测的端到端Speaker Diarization

End-to-End Speaker Diarization Conditioned on Speech Activity and Overlap Detection

Center for Language and Speech Processing, Johns Hopkins University(约翰霍普金斯大学语言与语音处理中心)

Yuki Takashima, Yusuke Fujita, Shinji Watanabe, Shota Horiguchi, Paola García, Kenji Nagamatsu

摘要:

在本文中,我们提出了一个有条件的多任务学习方法的端到端神经说话人日记(EEND)。与传统的基于聚类的方法相比,EEND系统表现出良好的性能,特别是在重叠语音的情况下。在本文中,为了进一步提高EEND系统的性能,我们提出了一种新的多任务学习框架,解决了扬声器日记和所需的子任务,同时明确考虑任务依赖。我们优化说话人日记的条件下,语音活动和重叠检测的子任务的说话人日记,基于概率链规则。实验结果表明,我们提出的方法可以利用一个子任务来有效地建模说话人日记,并优于传统的EEND系统的日记错误率。

英文摘要:

In this paper, we present a conditional multitask learning method for end-to-end neural speaker diarization (EEND). The EEND system has shown promising performance compared with traditional clustering-based methods, especially in the case of overlapping speech. In this paper, to further improve the performance of the EEND system, we propose a novel multitask learning framework that solves speaker diarization and a desired subtask while explicitly considering the task dependency. We optimize speaker diarization conditioned on speech activity and overlap detection that are subtasks of speaker diarization, based on the probabilistic chain rule. Experimental results show that our proposed method can leverage a subtask to effectively model speaker diarization, and outperforms conventional EEND systems in terms of diarization error rate.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑