搜索结果: 1-4 共查到“工学 GMM-UBM”相关记录4条 . 查询时间(0.109 秒)
将语种和说话人识别的方法应用到英语发音错误检测系统, 提出一种基于广义线性区分序列支持向量机 (Generalized linear discriminant sequence based SVM, GLDS-SVM)的发音错误检测方法. 主要创新点为: 1)提出一种基于状态拼接的特征规整方案, 增强SVM对发音特征的建模能力; 2)提出一种基于多模型融合的模型训练策略, 该策略可以更加充分地利用...
在说话人辨认任务中, Gau ss混合模型- 通用背景模型(Gau ssianm ix tu rem o de l2un ive r sa lbackg ro und m o de l, GMM2U BM) 采用帧向量进行建模和识别, 突出了说话人个性特征, 但受信道影响较大; 支持向量机(suppo r t vecto rm ach ine, SVM)利用帧向量在空间中分布的Gau ss混合的均...
Modeling High-level Information by Using Gaussian Mixture Correlation for GMM-UBM Based Speaker Recognition
Modeling High-level Information Gaussian Mixture Correlation GMM-UBM Based Speaker Recognition
2013/6/28
The Gaussian mixture model-universal background model (GMM-UBM) has been dominant in text-independent speaker recognition tasks. However the conventional GMM-UBM method assumes that each Gaussian mixt...
COMBINING SELECTION TREE WITH OBSERVATION REORDERING PRUNING FOR EFFICIENT SPEAKER IDENTIFICATION USING GMM-UBM
COMBINING SELECTION TREE WITH OBSERVATION REORDERING PRUNING EFFICIENT SPEAKER IDENTIFICATION GMM-UBM
2013/6/28
In this paper a new method of reducing the computational load for Gaussian Mixture Model Universal Background Model (GMM-UBM) based speaker identification is proposed. In order to speed up the selecti...