Abstract
In this paper, we introduce a new method to estimate the total variability space using sparse probabilistic Principal Component Analysis (PCA) with the Baum-Welch statistics for speaker verification. In conventional method, probabilistic PCA is used, which is a probabilistic formulation for PCA. Recently some methods improve interpretability by sparse representation through adding an L1 regularizer. We introduce a Laplacian prior to each element in the transformation matrix, since Laplacian prior is equivalent to L1 regularization. Variational inference is used and we can drive the EM algorithm formulas for estimating the space with the statistics. After WCCN, the cosine similarity scoring is used for decision. The experiments have been run on the NIST SRE 2008 data, and the results show that the performance can be improved 10.2% for female and is comparable for male.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Kenny, P., Gilles, B., Pierre, D.: Eigenvoice Modeling With Sparse Training Data. IEEE Trans. Speech and Audio Proc. 13(3), 345–354 (2005)
Kenny, P., et al.: Joint Factor Analysis Versus Eigenchannels in Speaker Recognition. IEEE Trans. Speech and Language Proc. 15(4), 1435–1447 (2007)
Dehak, N., et al.: Support vector machines versus fast scorring in the low-dimensional total variability space for speaker verification. In: Interspeech, Brighton, UK (2009)
Reynolds, D.A., Quatieri, T., Dunn, R.: Speaker verification using adapted Gaussian mixture models. Digital Signal Processing 10(1-3) (2000)
Tipping, M., Bishop, C.: Mixtures of probabilistic principal component analyzers. Neural Computation 11, 435–474 (1999)
Roweis, S.: EM Algorithms for PCA and SPCA. In: Advances in Neural Information Processing Systems (1998)
Jolliffe, I.T.: Principal Component analysis. Springer, New York (1986)
Guan, Y., Dy, J.G.: Sparse Probabilistic Principal Component Analysis. In: Proceedings of the 12th International Conference on Artificial Intelligence and Statistics, Clearwater Beach, Florida, USA, vol. 5 (2009)
Bishop, C.M.: Variational Principal Components. In: Proceedings Ninth International Conference on Artificial Neural Network, vol. 1, pp. 509–514
Li, M., et al.: Speaker Verification using Sparse Representations on Total Variability I-Vector. In: Interspeech, Florence, Italy (2011)
Hatch, A., Kajarekar, S., Stolcke, A.: Withinclass covariance normalization for svm-based speaker recognition. In: Interspeech – 9th International Conference on Spoken Language Processing-ICSLP, vol. 3, pp. 1471–1474 (2006)
Martinez, A.M., Kak, A.C.: PCA versus LDA. IEEE Transactions on Pattern Analysis and Machine Intelligence 23(2), 228–233 (2004)
Shum, S., et al.: Unsupervised Speaker Adaptation based on the Cosine Similarity for Text-Independent Speaker Verification, Odyssey. In: The Speaker and Language Recognition Workshop (2010)
Dehak, N., et al.: Cosine Similarity Scoring without Score Normalization Techniques, Odyssey. In: The Speaker and Language Recognition Workshop (2010)
Auckenthaler, R., Carey, M., Thomas, H.L.: Score Normalization for Text-Independent Speaker Verification Systems. Digital Signal Processing 10, 42–54 (2000)
Daniel, G.R., Carol, Y.E.W.: Joint Factor Analysis for Speaker Recognition reinterpreted as Signal Coding using Overcomplete dictionaries, Odyssey. In: The Speaker and Language Recognition Workshop (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lei, Z., Ye, J. (2012). Estimating the Total Variability Space Using Sparse Probabilistic Principal Component Analysis. In: Zheng, WS., Sun, Z., Wang, Y., Chen, X., Yuen, P.C., Lai, J. (eds) Biometric Recognition. CCBR 2012. Lecture Notes in Computer Science, vol 7701. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35136-5_45
Download citation
DOI: https://doi.org/10.1007/978-3-642-35136-5_45
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-35135-8
Online ISBN: 978-3-642-35136-5
eBook Packages: Computer ScienceComputer Science (R0)