XU Longting, YANG Zhen, SUN Linhui, “Simplification of I-Vector Extraction for Speaker Identification,” Chinese Journal of Electronics, vol. 25, no. 6, pp. 1121-1126, 2016, doi: 10.1049/cje.2016.10.016
Citation: XU Longting, YANG Zhen, SUN Linhui, “Simplification of I-Vector Extraction for Speaker Identification,” Chinese Journal of Electronics, vol. 25, no. 6, pp. 1121-1126, 2016, doi: 10.1049/cje.2016.10.016

Simplification of I-Vector Extraction for Speaker Identification

doi: 10.1049/cje.2016.10.016
Funds:  This work is supported by the National Natural Science Foundation of China (No.60971129, No.61271335, No.61501251), the Scientific Innovation Research Programs of College Graduate in Jiangsu Province (No.CXZZ13_0488), Key Laboratory of the Ministry of Public Security Smart Speech Technology (No.2014ISTKFKT02), the Natural Science Foundation of Jiangsu Province (No.BK20140891), the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (No.13KJB510020), and the Science Foundation of Nanjing University of Posts and Telecommunications (No.NY214191).
  • Received Date: 2015-06-08
  • Rev Recd Date: 2015-07-25
  • Publish Date: 2016-11-10
  • The identity vector (i-vector) approach has been the state-of-the-art for text-independent speaker recognition, both identification and verification in recent years. An i-vector is a low-dimensional vector in the so-called total variability space represented with a thin and tall rectangular matrix. This paper introduces a novel algorithm to improve the computational and memory requirements for the application. In our method, the series of symmetric matrices can be represented by diagonal expression, sharing the same dictionary, which to some extent is analogous to eigen decomposition, and we name this algorithm Eigen decomposition like factorization (EDLF). Similar algorithms are listed for comparison, in the same condition, our method shows no disadvantages in identification accuracy.
  • loading
  • N. Dehak, P. Kenny, R. Dehak, et al., "Front-end factor analysis for speaker verification", IEEE Transactions on Audio, Speech, and Language Processing, Vol.19, No.4, pp.788-798, 2011.
    S.J. Prince and J.H. Elder, "Probabilistic linear discriminant analysis for inferences about identity", Proc. of ICCV, Rio de Janeiro, Brazil, pp.1-8, 2007.
    P. Kenny, "Bayesian speaker verification with heavy-tailed priors", Proc. of Odyssey, Brno, Czech Republic, page 14, 2010.
    K.A. Lee, L. Anthony, C. You, et al., "Multisession plda scoring of i-vector for partially open-set speaker detection", Proc. of INTERSPEECH, Lyon, France, pp.3651-3655, 2013.
    S. Cumani and P. Laface, "Factorized sub-space estimation for fast and memory effective i-vector extraction", IEEE Transactions on Audio, Speech, and Language Processing, Vol.22, No.1, pp.248-259, 2014.
    O. Glembek, L. Burget, P. Matejka, et al., "Simplification and optimization of i-vector extraction", Proc. of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Prague, Czech Republic, pp.4516-4519, 2011.
    L. Xu, K.A. Lee, H. Li,, et al., "Sparse coding of total variability matrix", Proc. of INTERSPEECH, Dresden, Germany, pp.1022-1026, 2015.
    P. Kenny, G. Boulianne, and P. Dumouchel, "Eigenvoice modeling with sparse training data", IEEE Transactions on Audio, Speech, and Language Processing, Vol.13, No.3, pp.345-354, 2005.
    M. Senoussaoui, P. Kenny, N. Dehak, et al., "An i-vector extractor suitable for speaker recognition with both microphone and telephone speech", Proc. of Odyssey, Brno, Czech Republic, pp.6, 2010.
    A. Kanagasundaram, D. Dean, S. Sridharan, et al., "i-vector based speaker recognition using advanced channel compensation techniques", Computer Speech & Language, Vol.28, No.1, pp.121-140, 2014.
    R. Rubinstein, A.M. Bruckstein, and M. Elad, "Dictionaries for sparse representation modeling", Proceedings of the IEEE, Vol.98, No.6, pp.1045-1057, 2010.
  • 加载中

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Article Metrics

    Article views (424) PDF downloads(516) Cited by()
    Proportional views
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return