Partial Coupling of Optimal Transport for Spoken Language Identification
Partial Coupling of Optimal Transport for Spoken Language Identification
复制标题
DOI：
10.48550/arxiv.2203.17036
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Xugang Lu;Peng Shen;Yu Tsao;H. Kawai
中科院分区：
文献类型：
--
作者：
Xugang Lu;Peng Shen;Yu Tsao;H. Kawai
为了减少跨域口语识别系统中的域差异，提高系统的识别性能，提出了一种基于最优传输的联合分布对齐（JDA）模型。采用基于OT的差异度量方法对训练数据集和测试数据集进行JDA。在我们之前的研究中，假设训练集和测试集共享相同的标签空间。然而，在真实的应用中，测试集的标签空间只是训练集的标签空间的子集。为了分布对齐而完全匹配训练域和测试域可能会引入负域转移。本文提出了一种基于部分最优运输（POT）的JDA模型，在JDA期间仅允许OT的部分偶联。此外，由于测试数据的标签是未知的，在POT中，在域对齐期间，基于传输成本自适应地设置对耦合的软加权。在跨域SLID任务上进行了实验，以评估所提出的UDA。结果表明，我们提出的UDA显着提高了性能，由于部分耦合的OT的考虑。
In order to reduce domain discrepancy to improve the performance of cross-domain spoken language identification (SLID) system, as an unsupervised domain adaptation (UDA) method, we have proposed a joint distribution alignment (JDA) model based on optimal transport (OT). A discrepancy measurement based on OT was adopted for JDA between training and test data sets. In our previous study, it was supposed that the training and test sets share the same label space. However, in real applications, the label space of the test set is only a subset of that of the training set. Fully matching training and test domains for distribution alignment may introduce negative domain transfer. In this paper, we propose an JDA model based on partial optimal transport (POT), i.e., only partial couplings of OT are allowed during JDA. Moreover, since the label of test data is unknown, in the POT, a soft weighting on the coupling based on transport cost is adaptively set during domain alignment. Experiments were carried out on a cross-domain SLID task to evaluate the proposed UDA. Results showed that our proposed UDA significantly improved the performance due to the consideration of the partial couplings in OT.