国家自然科学基金(90820011)

作品数:8被引量:60H指数:3
导出分析报告
相关作者:刘文举徐波倪崇嘉李鹏廖逢钗更多>>
相关机构:中国科学院自动化研究所山东财政学院三明学院河南理工大学更多>>
相关期刊:《Chinese Journal of Acoustics》《Chinese Journal of Electronics》《声学学报》《Science China(Information Sciences)》更多>>
相关主题:MUSIC大词汇量模型自适应连续语音识别系统中文信息处理更多>>
相关领域:电子电信自动化与计算机技术交通运输工程机械工程更多>>
-

检索结果分析

结果分析中...
条 记 录,以下是1-8
视图:
排序:
Auditory filter based broadband MUSIC algorithm for sound source localization被引量:7
《Chinese Journal of Acoustics》2013年第4期439-453,共15页LIAO Fengchai LI Peng LIU Wenju 
supported by the National Nature Science Foundation of China(91120303,61273267,90820011);FuJian Nature Science Foundation(2009J01296)
Based on the analysis of the shortcomings of broadband MUSIC algorithm with short-time Fourier transform (SF-MUSIC) for sound source localization, a broadband MUSIC algorithm with auditory filter (AF-MUSIC) was pr...
关键词:MUSIC SF DOA 
汉语语音识别中基于音节的声学模型改进算法被引量:1
《计算机应用》2013年第6期1742-1745,共4页晁浩 杨占磊 刘文举 
国家自然科学基金资助项目(91120303,90820303,90820011);国家973计划项目(2004CB318105);国家863计划项目(20060101Z4073,2006AA01Z194)
针对汉语语音识别中协同发音现象引起的语音信号的易变性,提出一种基于音节的声学建模方法。首先建立基于音节的声学模型以解决音节内部声韵母之间的音变现象,并提出以音节内双音子模型来初始化基于音节声学模型的参数以缓解训练数据稀...
关键词:语音识别 协同发音 音变 声学建模 音节模型 
采用听觉滤波器的宽带MUSIC声源定位方法被引量:7
《声学学报》2012年第6期642-650,共9页廖逢钗 李鹏 刘文举 
国家自然科学基金项目(90820011,91120303,90820303);福建省自然科学基金项目(2009J01296)
在分析了采用短时傅里叶变换的宽带MUSIC声源定位算法(SF-MUSIC)存在问题的基础上,提出了一种采用听觉滤波器的宽带MUSIC声源定位算法(AF-MUSIC)。该算法使用听觉滤波器组对传声器阵列接收到的信号进行不等带宽分解后,在各个频率通道上...
关键词:MUSIC算法 声源定位 定位方法 滤波器 宽带 听觉 短时傅里叶变换 定位算法 
Integrating induced probability into decoding for large vocabulary continuous speech recognition被引量:2
《Chinese Journal of Acoustics》2012年第3期338-352,共15页YANG Zhanlei LIU Wenju CHAO Hao 
supported by the National Nature Science Foundation of China(91120303,90820011, 90820303);the 863 National High Technology Development Project of China(20060101Z4073,2006AA01Z194);the National Grand Fundamental Research 973 Program of China(2004CB318105)
This paper integrates location information of frames into conventional acoustic model (AM) and language model (LM) likelihoods, in order to distinguish potential path can- didates more precisely at decoding stage....
A signal subspace dimension estimator based on F-norm with application to subspace-based multi-channel speech enhancement被引量:2
《Chinese Journal of Acoustics》2012年第3期353-368,共16页LI Chao LIU Wenju 
supported by the National Nature Science Foundation of China(91120303,90820011, 90820303);the National Grand Fundamental Research 973 Program of China(2004CB318105)
Although the signal subspace approach has been studied extensively for speech enhancement, no good solution has been found to identify signal subspace dimension in multi- channel situation. This paper presents a signa...
Monaural voiced speech segregation based on elaborate harmonic grouping strategies
《Science China(Information Sciences)》2011年第12期2471-2480,共10页LIU WenJu ZHANG XueLiang JIANG Wei LI Peng XU Bo 
supported by the National Natural Science Foundation of China(Grant Nos.60675026,90820303,90820011)
In this paper, an enhanced algorithm based on several elaborate harmonic grouping strategies for monaural voiced speech segregation is proposed. Main achievements of the proposed algorithm lie in three aspects. Firstl...
关键词:computational auditory scene analysis voiced speech separation harmonistic principle minimum amplitude principle elaborate harmonic grouping strategies 
Unsupervised Learning of Gaussian Mixture Model with Application to Image Segmentation被引量:2
《Chinese Journal of Electronics》2010年第3期451-456,共6页LI Bo LIU Wenju DOU Lihua 
This work is supported by the National Natural Science Foundation of China (No.60675026, No.60121302, No.90820011), 863 China National High Technology Development Project (No.20060101Z4073, No.2006AA01Z194), the National Grand Fundamental Research 973 Program of China (No.2004CB318105).
Density estimation via Gaussian mixture modeling has been successfully applied to image segmentation, speech processing and other fields relevant to clustering analysis and Probability density function (PDF) modelin...
关键词:Gaussian mixture model Model order Degenerating model Elliptically contoured distributions Image segmentation. 
汉语大词汇量连续语音识别系统研究进展被引量:39
《中文信息学报》2009年第1期112-123,128,共13页倪崇嘉 刘文举 徐波 
国家重点基础研究发展计划(973)资助项目(2004CB318105);国家高技术研究发展计划(863)资助项目(2006AA01Z194,20060101Z4073);国家自然科学基金资助项目(60675026,60121302,90820011)
大词汇量连续语音识别(LVCSR)技术近年来发展迅速,并在许多领域得到了广泛的应用,国内外许多大公司加大了对语音识别技术的研究,不少商业化的语音识别系统已经面世,并得到较为广泛的使用。该文综述了近年来大词汇量连续语音识别技术的...
关键词:计算机应用 中文信息处理 综述 语音识别 模型自适应 搜索技术 
检索报告 对象比较 聚类工具 使用帮助 返回顶部