10.1631/jzus.C13e0320

,"/> 语音信号辅音/元音分割的快速全局模糊c均值聚类算法
Please wait a minute...
FITEE
Front. Inform. Technol. Electron. Eng.  2014, Vol. 15 Issue (7): 551-563    DOI: 10.1631/jzus.C1300320
    
语音信号辅音/元音分割的快速全局模糊c均值聚类算法
Xian Zang, Felipe P. Vista Iv, Kil To Chong
Department of Electronic Engineering, Jeonbuk National University, Jeonju-si 561-756, Korea; Advanced Electronics and Information Research Center, Jeonbuk National University, Jeonju-si 561-756, Korea
Fast global kernel fuzzy c-means clustering algorithm for consonant/vowel segmentation of speech signal
Xian Zang, Felipe P. Vista Iv, Kil To Chong
Department of Electronic Engineering, Jeonbuk National University, Jeonju-si 561-756, Korea; Advanced Electronics and Information Research Center, Jeonbuk National University, Jeonju-si 561-756, Korea
 全文: PDF 
摘要： 创新方法：传统的模糊c均值方法（FCM）有两个缺点：对初始值要求严格，无法处理非线性分割数据。通过使用基于核的模糊c均值聚类法（KFCM-F）作为本地搜索方法，采用渐进方法获得近乎最优的结果，这种方法的渐进性和KFCM-F的非线性，可以避免FCM的两个缺点。
研究手段：使用KFCM-F处理数据，在不显著影响实验结果的情况下，设计了一个加速计划以降低计算复杂度。采用非线性人工数据组和现实数据组作为语音信号，进行辅音/元音分割，以检测这种新算法的性能。
重要结论：KFCM-F方法巧妙地避免了传统FCM方法的两个缺点。我们设计的算法（FGKFCM-F）继承了KFCM-F和全局模糊c均值方法（GFCM）的优点，得以实现基于非线性分割数据组的近乎最优解。此外，我们设计的加速计划大大降低了整个计算的复杂度。实验结果证实，FGKFCM-F比其他方法更适合处理人工和现实数据。
关键词： 模糊c均值聚类法;  核方法;  全局优化;  辅音/元音分割    
Abstract: We propose a novel clustering algorithm using fast global kernel fuzzy c-means-F (FGKFCM-F), where F refers to kernelized feature space. This algorithm proceeds in an incremental way to derive the near-optimal solution by solving all intermediate problems using kernel-based fuzzy c-means-F (KFCM-F) as a local search procedure. Due to the incremental nature and the nonlinear properties inherited from KFCM-F, this algorithm overcomes the two shortcomings of fuzzy c-means (FCM): sensitivity to initialization and inability to use nonlinear separable data. An accelerating scheme is developed to reduce the computational complexity without significantly affecting the solution quality. Experiments are carried out to test the proposed algorithm on a nonlinear artificial dataset and a real-world dataset of speech signals for consonant/vowel segmentation. Simulation results demonstrate the effectiveness of the proposed algorithm in improving clustering performance on both types of datasets.
Key words: Fuzzy c-means clustering    Kernel method    Global optimization    10.1631/jzus.C13e0320

')" href="#">Consonant/vowel segmentation

An erratum to this article can be found at doi:10.1631/jzus.C13e0320
收稿日期: 2013-11-02 出版日期: 2014-07-10
CLC:  TN912  
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　
Xian Zang
Felipe P. Vista Iv
Kil To Chong

引用本文:

Xian Zang, Felipe P. Vista Iv, Kil To Chong. Fast global kernel fuzzy c-means clustering algorithm for consonant/vowel segmentation of speech signal. Front. Inform. Technol. Electron. Eng., 2014, 15(7): 551-563.

链接本文:

http://www.zjujournals.com/xueshu/fitee/CN/10.1631/jzus.C1300320    或    http://www.zjujournals.com/xueshu/fitee/CN/Y2014/V15/I7/551

[1] Sheng-kai Yang, Jian-yi Meng, Hai-bin Shen. 基于邻域子空间缩放维持局部线性的原像算法[J]. Front. Inform. Technol. Electron. Eng., 2014, 15(4): 254-264.