###

DOI:

计算机系统应用英文版:2011,20(12):127-131

View/Add Comment 过刊浏览高级检索 HTML

←前一篇 | 后一篇→

码上扫一扫！

下载全文

基于状态持续时间的HMM语音识别模型

孙玉莎, 王朝立, 白洁, 鲁国辉

(上海理工大学光电与计算机工程学院,上海 200093)

Speech Recognition Models Based on State Duration Hmm

SUN Yu-Sha, WANG Chao-Li, BAI Jie, LU Guo-Hui

(School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China)

摘要

图/表

参考文献

相似文献

本文已被：浏览 1512次下载 4107次
Received:May 30, 2011 Revised:June 24, 2011

中文摘要: 针对经典隐马尔可夫模型对状态持续时间的函数表达与实际语音的物理事实不相符合这一缺点,在通常隐马尔可夫的基础上引入状态持续时间参数,建立基于状态持续时间的HMM 语音识别模型（SDHMM),并用其进行语音识别实验,与经典隐马尔可夫模型相比,识别率有所提高。

中文关键词: 隐马尔可夫模型状态持续时间语音识别

Abstract:In order to solve the Function expression of state duration of Classic hidden Markov model doesn't match The physical facts of actual voice, this article proposed a state duration based HMM（SDHMM), bring the state duration parameters into the usual hidden Markov model. With its voice recognition experiments with the classical hidden Markov models, the recognition rate improved.

keywords: Hidden Markov model state duration speech recognition

文章编号： 中图分类号： 文献标志码：

基金项目:

引用文本：
孙玉莎,王朝立,白洁,鲁国辉.基于状态持续时间的HMM语音识别模型.计算机系统应用,2011,20(12):127-131
SUN Yu-Sha,WANG Chao-Li,BAI Jie,LU Guo-Hui.Speech Recognition Models Based on State Duration Hmm.COMPUTER SYSTEMS APPLICATIONS,2011,20(12):127-131

Author Name	Affiliation
SUN Yu-Sha	School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
WANG Chao-Li	School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
BAI Jie	School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
LU Guo-Hui	School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China