[1]于云,周伟栋. 基于压缩感知的鲁棒性说话人识别参数研究[J].计算机技术与发展,2016,26(03):18-22.
　YU Yun,ZHOU Wei-dong. Research on Robust Speaker Recognition Parameters Based on Compressed Sensing[J].,2016,26(03):18-22.
点击复制

 基于压缩感知的鲁棒性说话人识别参数研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
26
期数:
2016年03期
页码:
18-22
栏目:
智能、算法、系统工程
出版日期:
2016-03-10

文章信息/Info

Title:
 Research on Robust Speaker Recognition Parameters Based on Compressed Sensing
文章编号:
1673-629X（2016)03-0018-05
作者:
 于云; 周伟栋
 南京邮电大学 通信与信息工程学院
Author(s):
 YU Yun; ZHOU Wei-dong
关键词:
 ; 压缩感知; 谱减法; 特征参数; 鲁棒性
Keywords:
 ; compressed sensing; spectral subtraction; feature parameters; robustness
分类号:
TN912.3
文献标志码:
A
摘要:
 奈奎斯特采样下的说话人识别,当为了确保高的识别率而采集较长时间说话人语音时,采样数据量特别大,其中有许多冗余造成了采样资源的浪费,压缩感知理论可以很好地解决此问题。基于压缩感知理论,文中利用行阶梯观测矩阵对信号进行投影,研究了压缩比与识别率的关系,在压缩比为1：2时,保证识别率的同时,使得采样数据量减少为原来的一半。在有噪环境下,将谱减法运用到压缩感知和特征提取过程中,在无需重构时域信号的前提下,直接从已估计的干净语音功率谱中提取具有鲁棒性的特征参数CS-SSMFCC( Compressed Sensing Spectral Subtraction Mel Frequency Cepstral Co-efficient)。实验结果表明,与传统的识别参数MFCC( Mel Frequency Cepstral Coefficient)相比,CS-SSMFCC可以有效地提高系统的鲁棒性,具有很好的抗噪性能。
Abstract:
 Speaker recognition under Nyquist sampling has got a large amount of data in order to ensure a high recognition rate,resulting in a waste of sampling resources,and compressive sensing theory can solve this problem. Based on compressed sensing theory,it makes use of ladder observation matrix projection in this paper. When the compression ratio is 1:2,the system ensures the recognition rate,so that the sample data is reduced to half. Under noisy environment,spectral subtraction is applied in compressed sensing and feature extrac-tion,and feature parameters are extracted directly from estimated clean speech power spectrum CS-SSMFCC (Compressed Sensing Spec-tral Subtraction Mel Frequency Cepstral Coefficient) . Experimental results show that compared with the traditional identification parame-ter MFCC (Mel frequency Cepstral Coefficient),CS-SSMFCC based on spectral subtraction under CS framework can effectively im-prove the robustness of the system,with good anti-noise performance.

相似文献/References:

[1]张爱华 薄禄裕 盛飞 杨培.基于小波变换的压缩感知在图像加密中的应用[J].计算机技术与发展,2011,(12):145.
　ZHANG Ai-hua,BO Lu-yu,SHENG Fei,et al.Compressed Sensing Based on Single Layer Wavelet Transform for Image Encryption[J].,2011,(03):145.
[2]王韦刚 庄伟胤.基于NIOS Ⅱ的图像压缩感知[J].计算机技术与发展,2012,(04):12.
　WANG Wei-gang,ZHUANG Wei-yin.Compressed Sensing of Image Based on NIOS Ⅱ[J].,2012,(03):12.
[3]王韦刚 胡海峰.基于压缩感知的协作频谱检测[J].计算机技术与发展,2012,(12):241.
　WANG Wei-gang,HU Hai-feng.Collaborative Spectrum Detection Based on Compressed Sensing[J].,2012,(03):241.
[4]张晓咏,熊承义,胡开云,等.基于灰度纹理信息的图像压缩感知编码与重构[J].计算机技术与发展,2013,(01):47.
[5]刘洋,季薇,侯晓赟.一种改进的基于 OMP 重建的宽带频谱感知算法[J].计算机技术与发展,2013,(01):99.
　LIU Yang,JI Wei,HOU Xiao-yun.A Modified Spectrum Sensing Algorithm for Wideband Cognitive Radio Based on OMP[J].,2013,(03):99.
[6]彭钰,侯晓赟.基于二维压缩感知的双选信道估计[J].计算机技术与发展,2013,(10):220.
　PENG Yu,HOU Xiao-yun.Doubly Selective Channel Estimation Based on Two Dimension Compressed Sensing[J].,2013,(03):220.
[7]李熔.基于截尾估计的概率估计方法[J].计算机技术与发展,2014,24(02):101.
　LI Rong.Probability Estimation Method Based on Truncated Estimation[J].,2014,24(03):101.
[8]李燕,王博.基于压缩感知的数据压缩与检测[J].计算机技术与发展,2014,24(03):198.
　LI Yan,WANG Bo.Data Compression and Detection Based on Compressive Sensing[J].,2014,24(03):198.
[9]周飞飞,李雷.小波高频子带变换裁剪阈值SAMP算法研究[J].计算机技术与发展,2014,24(05):83.
　ZHOU Fei-fei,LI Lei.Research on Clipping Threshold SAMP Algorithm Based on High Frequency Sub-band Wavelet Transform[J].,2014,24(03):83.
[10]刘正其,季薇.一种改进的基于BOMP的宽带频谱感知算法[J].计算机技术与发展,2014,24(06):118.
　LIU Zheng-qi,JI Wei.A Modified Spectrum Sensing Algorithm for Wideband Cognitive Radio Based on BOMP[J].,2014,24(03):118.
[11]徐志坚,邱晓晖. 采用压缩感知的协作多点信道反馈算法研究[J].计算机技术与发展,2014,24(10):221.
　XU Zhi-jian,QIU Xiao-hui. Study on Channel Feedback Algorithm Using Compressed Sensing for Coordinated Multiple Point[J].,2014,24(03):221.
[12]柯家龙,李继楼. 压缩感知中的投影矩阵优化算法[J].计算机技术与发展,2015,25(03):95.
　KE Jia-long,LI Ji-lou. Algorithm of Optimization for Projection Matrix in Compressive Sensing[J].,2015,25(03):95.
[13]李尚靖[],朱琦[][],朱俊华[]. 基于压缩感知和正弦字典的语音编码新方案[J].计算机技术与发展,2015,25(04):188.
　LI Shang-jing[],ZHU Qi[][],ZHU Jun-hua[]. A New Scheme of Speech Coding Based on Compressed Sensing and Sinusoidal Dictionary[J].,2015,25(03):188.
[14]郭海亮. 基于GEP算法的压缩感知语音观测序列建模[J].计算机技术与发展,2015,25(05):46.
　GUO Hai-liang. Speech Signals Measurements Sequence Modeling in Compressed Sensing Based on GEP[J].,2015,25(03):46.
[15]郭青青,李雷. 基于SiT-ROMP算法的视频封装帧压缩重构研究[J].计算机技术与发展,2015,25(08):113.
　GUO Qing-qing,LI Lei. Research on Compressing and Reconstructing of Encapsulated Video Frame Based on Self-iterative Threshold ROMP Algorithm[J].,2015,25(03):113.
[16]李继楼,柯家龙. 基于压缩感知的WSN数据压缩与重构[J].计算机技术与发展,2015,25(09):111.
　LI Ji-lou,KE Jia-long. Data Compression and Recovery of WSN Based on Compressive Sensing[J].,2015,25(03):111.
[17]钱阳,李雷. 一种基于新型KPCA算法的视频压缩感知算法[J].计算机技术与发展,2015,25(10):101.
　QIAN Yang,LI Lei. A Video Compressed Sensing Algorithm Based on Novel KPCA[J].,2015,25(03):101.
[18]玲玲,齐丽娜. 特征字典与自适应联合的BCS-UWB信道估计[J].计算机技术与发展,2015,25(12):195.
　WANG Ling-ling,QI Li-na. Ultra-wideband Channel Estimation Based on Bayesian Compressive Sensing of Eigen-based Dictionary and Adaptive Joint[J].,2015,25(03):195.
[19]孙君,孙照伟. 基于压缩感知的信道互易性补偿方法[J].计算机技术与发展,2015,25(12):210.
　SUN Jun,SUN Zhao-wei. A Compensation Method for Channel Non-reciprocity Based on Compressive Sensing[J].,2015,25(03):210.
[20]陈天宇,吴凡,马世杰,等. 基于CS和LS-SVM的入侵检测算法[J].计算机技术与发展,2016,26(05):99.
　CHEN Tian-yu,WU Fan,MA Shi-jie,et al. Intrusion Detection Algorithm Based on Compressed Sensing and Least Square Support Vector Machine[J].,2016,26(03):99.

更新日期/Last Update: 2016-05-24