The server is under maintenance between 08:00 to 12:00 (GMT+08:00), and please visit later.
We apologize for any inconvenience caused
Login
|
Sign Up
|
Oriprobe Inc.
|
Feed
Home
Journals
Order
TOC Alerts
Subscription
Products & Services
Pricing
FAQ
About
Journal Articles
Laws/Policies/Regulations
Companies/Products
Title, abstract, keywords:
Combined Search
Advanced Search
Pay per View through On Demand Search
Package:
ALL
Astro-Earth Science
Agriculture
Physics
Mathematics
Arts & Humanities
Medline Collection
Health/Medicine/Biology
Chemistry/Chemical Engineering
CAOD
English Journals
Traditional Chinese Medicine
NPC CPPCC Journals
China Defense and Military Sciences
Author:
Journal / Book Title:
Year:
Volume:
Issue:
A novel WCCN clustering kernel applied in speaker recognition
Author(s):
XING Yu-juan
,
LI Heng-jie
,
HU Jian-jun
,
WANG Wan-jun
Pages:
167
-
172
Year:
2013
Issue:
2
Journal:
Journal of Yunnan University(Natural Sciences Edition)
Keyword:
语音识别
;
GMM超向量
;
Bhattacharyya距离
;
类内协方差归一化
;
支持向量机
;
Abstract:
针对说话人确认系统中GMM超向量建模计算复杂度高以及易受信道干扰的问题,提出一种新型的基于Bhattacharyya距离聚类的WCCN序列核函数算法.首先计算话者GMM模型之间的Bhattacharyya距离,根据该Bhattacharyya距离对话者模型进行聚类,得到聚类中心模型;紧接着对聚类中心模型的均值向量进行MAP自适应,进而生成超向量序列核函数;最后采用WCCN平滑归一化技术对序列核函数进行信道补偿,抑制噪音和信道畸变对核函数的影响.将该Bhattacharyya聚类WCCN核函数应用到SVM说话人确认系统,仿真实验结果表明该核函数可以有效地提高系统的识别准确率和识别速度.
Citations
No citation information
Related Articles
loading...