arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.02007eess.AS
收藏

用于说话人归属ASR的SOT触发神经聚类方法

SOT Triggered Neural Clustering for Speaker Attributed ASR

Xianrui Zheng, Guangzhi Sun, Chao Zhang, Philip C. Woodland

AI总结:

本文提出基于段级判别神经聚类(SDNC)的并行说话人归属ASR方法,无需额外非神经聚类即可分配说话人标签,并在AMI数据集上相比谱聚类显著降低DER和cpWER。

摘要:

本文介绍了一种利用神经聚类方法进行说话人归属ASR转录的新方法。通过并行处理机制,说话人分离(diarisation)和ASR可以同时进行,有助于防止级联系统中误差从一个子系统累积到下一个子系统。这是通过使用以序列化输出训练方法训练的ASR,并结合段级判别神经聚类(SDNC)来分配说话人标签实现的。借助SDNC,我们的系统不需要额外的非神经聚类方法来分配说话人标签,从而使整个系统能够完全基于神经网络。在AMI会议数据集上的实验结果表明,SDNC在AMI Eval集上的说话人分离错误率(DER)比谱聚类(SC)相对降低了19%。与使用SC的级联系统相比,使用SDNC的并行系统在Dev/Eval集上的cpWER分别相对改善了7%/4%。

英文摘要:

This paper introduces a novel approach to speaker-attributed ASR transcription using a neural clustering method. With a parallel processing mechanism, diarisation and ASR can be applied simultaneously, helping to prevent the accumulation of errors from one sub-system to the next in a cascaded system. This is achieved by the use of ASR, trained using a serialised output training method, together with segment-level discriminative neural clustering (SDNC) to assign speaker labels. With SDNC, our system does not require an extra non-neural clustering method to assign speaker labels, thus allowing the entire system to be based on neural networks. Experimental results on the AMI meeting dataset demonstrate that SDNC outperforms spectral clustering (SC) by a 19% relative diarisation error rate (DER) reduction on the AMI Eval set. When compared with the cascaded system with SC, the parallel system with SDNC gives a 7%/4% relative improvement in cpWER on the Dev/Eval set.

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。

↑