基于动态特性的D-LTSV语音端点检测方法

doi:10.3969/j.issn.1000-3428.2014.12.052

计算机工程 ›› 2014, Vol. 40 ›› Issue (12): 277-281.doi: 10.3969/j.issn.1000-3428.2014.12.052

基于动态特性的D-LTSV语音端点检测方法

赵欢,冯璐,陈佐,张希翔

湖南大学信息科学与工程学院,长沙 410082

收稿日期:2013-11-14 修回日期:2014-01-28 出版日期:2014-12-15 发布日期:2015-01-16
作者简介:赵欢(1967-),女,教授、博士生导师,主研方向:语音信号处理,嵌入式系统设计,嵌入式语音识别;冯璐,硕士研究生;陈佐,讲师、博士;张希翔,博士研究生。
基金项目:
国家自然科学基金资助面上项目(61173106)。

D-LTSV Voice Activity Detection Method Based on Dynamic Feature

ZHAO Huan,FENG Lu,CHEN Zuo,ZHANG Xixiang

College of Computer Science and Electronic Engineering,Hunan University,Changsha 410082,China

Received:2013-11-14 Revised:2014-01-28 Online:2014-12-15 Published:2015-01-16

摘要/Abstract

摘要： 端点检测是语音信号处理的一个关键环节。为提高语音在低性噪比以及非平稳噪声环境下的端点检测性能,在长时信号变化特征(LTSV)的基础上提出一种新的D-LTSV语音端点检测方法。采用Bartlett-Welch方法估计语音谱,分析语音谱在长时域上的熵,利用倒谱的动态特性分析方法提取连续帧熵值的动态变化特征。实验结果表明,D-LTSV综合考虑了语音的非平稳性和帧间非平稳性的动态变化情况,具有比LTSV更好的分辨能力,特别是在低性噪比和非平稳噪声的环境下,D-LTSV的分辨能力提升了50.77%,能够准确地进行端点检测,具有更强的鲁棒性。

关键词: 语音端点检测, 语音谱, 长时特征, 动态特性, 熵, 分辨力

Abstract: Voice Activity Detection(VAD) is a critical step for speech processing.In order to improve the performance of VAD in low Signal-to-noise Ratio(SNR) and nonstationary noise,this paper proposes a novel D-Long-term Signal Variability(LTSV) method based on LTSV for VAD.It uses the Bartlett-Welch method to estimate the signal spectrum,analyzes the entropy on the signal spectrum,and utilizes the analytical method of dynamic features used in the cepstrum to extract dynamic features of the entropy.D-LTSV takes into account the degree of nonstationarity of the signal and the dynamic changes between the frames.Compared with LTSV,experimental result shows that D-LTSV owns more discriminative power which is improved by 50.77 percent in low SNR and nonstationary noise and makes VAD more robust and more accurate.

Key words: Voice Activity Detection(VAD), voice spectrum, long-term feature, dynamic characteristic, entropy, discriminative power

中图分类号:

TP391

赵欢,冯璐,陈佐,张希翔. 基于动态特性的D-LTSV语音端点检测方法[J]. 计算机工程, 2014, 40(12): 277-281.

ZHAO Huan,FENG Lu,CHEN Zuo,ZHANG Xixiang. D-LTSV Voice Activity Detection Method Based on Dynamic Feature[J]. Computer Engineering, 2014, 40(12): 277-281.

参考文献

[1] Ghosh P K,Tsiartas A,Narayanan S.Robust Voice Activity Detection Using Long-term Signal Variability[J].IEEE Transactions on Audio,Speech,and Language Processing,2011,19(3):600-613.
[2] 朱恒军,于泓博,王发智.小波分析和支持向量机相融合的语音端点检测算法[J].计算机科学,2012,39(6):244-246.
[3] 于迎霞,史家茂.一种改进的基于倒谱特征的带噪端点检测方法[J].计算机工程,2004,30(19):85-87.
[4] Mcclellan S A,Gibson J D.Variable Rate CELP Based on Subband Flatness[C]//Proceedings of IEEE Inter-national Conference on Communications.Seattle,USA:IEEE Press,1995:1409-1413.
[5] Fuijimoto M,Ishizuka K,Nakatari T.A Voice Activity Detection Based on the Adaptive Integration of Multiple Speech Features and a Signal Decision Scheme[C]//Proceedings of Acoustics,Speech and Signal Con-ference.Las Vegas,USA:[s.n.],2008:4441-4444.
[6] Li Y,Zhao Y.Recognizing Emotions in Speech using Short-term and Long-term Features[C]//Proceedings of Spoken Language Conference.Urbana,USA:[s.n.],1998:2255-2258.
[7] Friedland G,Vinyals O,Huang Y,et al.Prosodic and Other Long-term Features for Speaker Diarization[J].IEEE Transactions on Audio,Speech,and Language Processing,2009,17(5):985-993.
[8] Fukuda T,Ichikawa O,Nishimura M.Short-and Long-term Dynamic Features for Robust Speech Recog-nition[C]//Proceedings of the 9th Annual Conference of the International Speech Communication Association.Sydney,Australia:[s.n.],2008:2262-2265.
[9] Ramrez J,Segura J C,Benítez C,et al.Efficient Voice Activity Detection Algorithms Using Long-term Speech Information[J].Speech Communication,2004,42(3):271-287.
[10] Fukuda T,Ichikawa O,Nishimura M.Long-term Spectro-temporal and Static Harmonic Features for Voice Activity Detection[J].Selected Topics in Signal Pro-cessing,2010,4(5):834-844.
[11] Fukuda T,Ichikawa O,Nishimura M.Phone-duration-dependent Long-term Dynamic Features for a Stochastic Model-based Voice Activity Detection [C]//Pro-ceedings of the 9th Annual Conference of the Inter-national Speech Communication Association.Sydney,Australia:[s.n.],2008:1293-1296.
[12] Khoa P C.Noise Robust Voice Activity Detection[D].Singapore:Nanyang Technological University,2012.

[1]	王旭, 陈永乐, 王庆生, 陈俊杰. 结合特征选择与集成学习的密码体制识别方案[J]. 计算机工程, 2021, 47(1): 139-145,153.
[2]	刘宇航, 马慧芳, 刘海姣, 余丽. 一种可重叠子空间K-Means聚类算法[J]. 计算机工程, 2020, 46(8): 58-63,71.
[3]	刘颖, 王聪, 苑迎, 蒋国佳, 刘珂祯, 王翠荣. 基于Pareto熵的多目标虚拟网络映射算法[J]. 计算机工程, 2020, 46(8): 153-159.
[4]	马鹏, 王泽宇, 钟卫东, 王绪安. 基于改进小波包分解的相关功耗攻击降噪方法[J]. 计算机工程, 2020, 46(7): 129-135,142.
[5]	安敬民, 李冠宇. 基于图熵极值理论的领域概念聚类方法[J]. 计算机工程, 2020, 46(6): 88-93.
[6]	沈国良, 翟江涛, 戴跃伟. 基于Markov模型的HTTP参数排序隐蔽信道检测方法[J]. 计算机工程, 2020, 46(2): 154-158,169.
[7]	赵军, 朱荽, 杨雯璟, 许彦辉, 庞宇. 一种基于密度峰值聚类的图像分割算法[J]. 计算机工程, 2020, 46(2): 274-278,285.
[8]	张俊, 张安康, 王辉. 基于SQAG模型的攻击熵优化算法[J]. 计算机工程, 2020, 46(10): 143-150.
[9]	刘佳美, 徐巧枝. 基于机器学习的SDN网络流量预测与部署策略[J]. 计算机工程, 2020, 46(10): 223-230.
[10]	雷翔霄,欧阳红林,肖乐意,范朝冬. 基于等价三维熵与鲸鱼优化算法的图像分割研究[J]. 计算机工程, 2019, 45(4): 217-222.
[11]	杨晨晨,马春梅,朱金奇. 基于智能手机的跌倒行为识别算法研究[J]. 计算机工程, 2019, 45(2): 178-183.
[12]	张文坤,汪西原,宋佳乾. 基于分数阶微分差与高斯曲率滤波的边缘检测算法[J]. 计算机工程, 2019, 45(2): 213-219.
[13]	蔡强, 李晶, 郝佳云. 基于聚焦损失与残差网络的远程监督关系抽取[J]. 计算机工程, 2019, 45(12): 166-170.
[14]	王燕, 段亚西. 基于局部熵拟合能量与全局信息的活动轮廓模型[J]. 计算机工程, 2019, 45(12): 207-213,221.
[15]	谭阳红, 张婧, 李肖. 基于复杂网络理论的电网节点重要度评估[J]. 计算机工程, 2019, 45(11): 281-286,297.

基于动态特性的D-LTSV语音端点检测方法

D-LTSV Voice Activity Detection Method Based on Dynamic Feature

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10