Full-covariance UBM and heavy-tailed PLDA in i-vector speaker verification
Full-covariance UBM and heavy-tailed PLDA in i-vector speaker verification
复制标题
DOI：
10.1109/icassp.2011.5947436
复制
发表时间：
2011-05
期刊：
影响因子：
--
通讯作者：
P. Matejka;O. Glembek;Fabio Castaldo;Md. Jahangir Alam;Oldrich Plchot;P. Kenny;L. Burget;J. Černocký
中科院分区：
文献类型：
--
作者：
P. Matejka;O. Glembek;Fabio Castaldo;Md. Jahangir Alam;Oldrich Plchot;P. Kenny;L. Burget;J. Černocký
本文介绍了基于i向量的说话人验证的最新进展。建议使用具有全协方差矩阵的通用背景模型（UBM），并进行了彻底的实验测试。使用简单的余弦距离和先进的技术，如概率线性判别分析（PLDA）和PLDA的重尾变体（PLDA- ht）对i向量进行评分。最后，在进入PLDA-HT建模之前，我们研究了i向量的降维。结果具有很强的竞争力：在NIST 2010 SRE任务上，单个全协方差LDA-PLDA-HT系统的结果接近复杂融合系统的结果。
In this paper, we describe recent progress in i-vector based speaker verification. The use of universal background models (UBM) with full-covariance matrices is suggested and thoroughly experimentally tested. The i-vectors are scored using a simple cosine distance and advanced techniques such as Probabilistic Linear Discriminant Analysis (PLDA) and heavy-tailed variant of PLDA (PLDA-HT). Finally, we investigate into dimensionality reduction of i-vectors before entering the PLDA-HT modeling. The results are very competitive: on NIST 2010 SRE task, the results of a single full-covariance LDA-PLDA-HT system approach those of complex fused system.