期刊文献+

基于独立分量分析和矢量量化的说话人识别 被引量:1

Speaker recognition system based on ICA and VQ
下载PDF
导出
摘要 使用独立分量分析(ICA)来提取说话人特征并与矢量量化(VQ)判决方法相结合,实现了一个高性能的基于ICA特征的VQ(ICA-VQ)说话人识别系统。通过ICA变换得到说话人语音特征基函数系数用于生成VQ码书,并导出包含能量失真的ICA-VQ码书失真测度和质心确定条件,生成最终的判决。仿真实验中ICA提取的特征分别用于不同系统实现说话人确认任务,各系统的DET曲线对比验证了VQ方法用于ICA特征分类判决的优势,同时不同码书尺寸下的等差率(EER)对比证明了VQ码书设计的有效性。 The paper combined the speaker feature extracted by ICA with VQ technique to the ICA-VQ speaker recognition system with high performance. A speaker speech ICA synthesis model was presented to get the speaker speech feature bases with ICA algorithm, and the coefficients of the bases were used in designing codebooks. A novel distortion measurement including energy and a new centroid condition were given. In the simulation experiment of speaker verification, the EER contrast results of VQ with different sizes prove that VQ codebooks are efficient and the DET cures of various methods show that VQ is a more suitable method to speaker recognition with the coefficients of ICA feature bases.
作者 屈微 刘贺平
出处 《计算机应用》 CSCD 北大核心 2005年第10期2401-2403,共3页 journal of Computer Applications
基金 国家十五科技攻关课题(2004BA616A1103)
关键词 独立分量分析(ICA) 矢量量化(VQ) 说话人识别 失真测度 Independent Component Analysis (ICA) Vector Quantization (VQ) speaker recognition distortion measurement
  • 引文网络
  • 相关文献

参考文献11

  • 1JANG GJ, YUN SJ, OH YH. Feature vector transformation using ICA and its application to speaker verification [ J]. Eurospeech,1999:767 - 770.
  • 2JANG GJ, LEE TW, OH YH. Learning statistically efficient features for speaker recognition [ A]. In proceedings ICASSP[C],2001.
  • 3CAO XR, LIU RW. General approach to blind source separation[J]. IEEE Trans. Signal Processing 1996, 78(4): 753 -766.
  • 4DUDA R, HART P, STORK D. Pattern Classification [ M]. Seconded. Wiley Interscience, New York, 2000.
  • 5REYNOLDS D, QUATIERI T, DUNN R. Speaker verification using adapted gaussian mixture models [J]. Digital Signal Processing 2000, 10 (1): 19 -41.
  • 6SONMEZ, M, HECK L, WEINTRAUB M, et al. A lognormal tied mixture model of pitch for prosody-based speaker recognition [ A].In Proc. 5th European Conference on Speech Communication and Technology, Eurospeech ( Rhodos, Greece) [C], 1997. 1391 -1394.
  • 7GERSHO A, GRAY R. Vector quantization and signal compression[M]. Kluwer Academic Publishers, Boston, 1991.
  • 8OLSHAUSEN BA, FIELD DJ. Emergence of simple - cell receptive Geld properties by learning a sparse code for natural images [ J].Nature, 1996(381): 607 -609.
  • 9BELL A J, SEJNOWSKI TJ. The independent components of natural scenes are edge filters[J]. Vision Research, 3 1997, 7(23): 3327-3338.
  • 10ROSCA J, KOFMEHL A. Cepstrum-like ICA representations for text independent speaker recognition [ A]. In proceedings ICASSP[C],2003.

同被引文献6

  • 1Ozkurt T E, Akgul T. Robust Text-independent speaker identification using bispectrum slice, signal processing and communications applications conference[A]. 2004. Proceedings of the IEEE 12th[C]. 2004, 418-421.
  • 2Deng J, Zheng T F, Song Z J, et al. Using predictive differential power spectrum and subband mel-spectrum centroid for robust speaker recognation in stationary noises[A]. The 4th International Conference on Machine Learning and Cybernetics[C]. 2005, 8: 4846-4851.
  • 3Hyvarinen A. New approximatioils of differential entropy for independent component analysis and projection pursuit [J]. Advance in Neural Information Processing Systems, 1998, 10: 273-279.
  • 4Hyvarinen A. Fast and robust tided-point algorithm for independent component analysis[J]. IEEE Trans on Neural Networks, 1999, 10(3): 626-634.
  • 5Jang Gil-Jin, Yun Seong-Jin, Oh Yung-Hwan. Feature vector transformation using independent component analysis and its application to speaker identification[J]. Euro Speech, 1999, 7(6): 767-770.
  • 6邱作春,曾庆宁.基于自适应波束形成和ICA的消噪系统[J].声学技术,2008,27(1):119-125. 被引量:1

引证文献1

二级引证文献2

;
使用帮助 返回顶部