Abstract
The problem of finding the intrinsic dimension of speech is addressed in this paper. Astructured vector quantization lattice, Self-Organizing Map (SOM), is used as a projection space for the data. The goal is to find a hypercubical SOM lattice where the sequences of projected speech feature vectors form continuous trajectories. The effect of varying the dimension of the lattice is investigated using feature vector sequences computed from the TIMIT database.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Aho, A., Hopcroft, J. and Ullman, J.: Data structures and algorithms. Addison-Wesley, 1983.
Bauer, H. and Pawelzik, K.: Quantifying the neighborhood preservation of self-organizing feature maps, IEEE Transactions on Neural Networks, 3(4) (1992), 570–579.
Bauer, H. and Villmann, T.: Growing a hypercubical output space in a self-organizing feature map, IEEE Transactions on Neural Networks, 8(2) (1997), 218–226.
Bishop, C., Hinton, G. and Strachan, I.: GTM through time. Technical Report 005, Neural Computing Research Group, Aston University, Birmingham, UK, 1997.
Bishop, C., Svensén, M. and Williams, C.: GTM: the generative topographic mapping, Neural Computation, 10(1) (1998), 215–234.
Camastra, F. and Vinciarelli, A.: Intrinsic dimension estimation of data: an approach based on Grassberger-Procaccia's algorithm, Neural Processing Letters, 14(1) (2001), 27–34.
Demartines, P. and Hérault, J. Curvilinear component analysis: A self-organizing neural network for nonlinear mapping of data sets, IEEE Transactions on Neural Networks, 8(1) (1997), 148–154.
Frisone, F., Firenze, F. and Morasso, P.: Application of topology-representing networks to the estimation of the intrinsic dimensionality of data. In Proceedings of ICANN'95, pages 323–327, Paris, France, 1995.
Grassberger, P. and Procaccia, I.: Measuring the strangeness of strange attractors, Physica, 9D (1983), 189–208.
Hecht-Nielsen, R.: Replicator neural networks for universal optimal source coding, Science, 269(5232) (1995), 1860–1863.
Kaski, S. and Lagus, K.: Comparing self-organizing maps. In Proceedings of ICANN'96, pages 809–814, Bochum, Germany, 1996.
Kohonen, T.: Self-organized formation of topologically correct feature maps, Biological Cybernetics, 43 (1982), 59–69.
Kohonen, T.: Self-Organizing Maps. Springer, Berlin (3rd extended ed. 2001), 1995.
Kokkonen, M.: Koartikulaatioilmiöiden mallittaminen itseorganisoituvan piirrekartan topologian avulla. Master's thesis, Helsinki University of Technology, Finland, 1991. In Finnish.
Lee, J., Lendasse, A. and Verleysen, M.: Curvilinear Distance Analysis versus Isomap. In Proceedings of ESANN'2002, pages 185–192, 2002.
Linde, Y., Buzo, A. and Gray, R.: An algorithm for vector quantizer design, IEEE Transactions on Communications, 28(1) (1980), 84–95.
Martinetz, T. and Schulten, K.: Topology representing networks, Neural Networks, 7(3) (1994), 507–522.
Nix, D. and Hogden, J.: Maximum-likelihood continuity mapping (MALCOM): An alternative to HMMs. In Proceedings of NIPS'11, pages 744–751, 1999.
Roweis, S.: Constrained hidden Markov models. In Proceedings of NIPS'12, pages 782–788, 2000.
Sammon, J. Jr.: A nonlinear mapping for data structure analysis, IEEE Transactions on Computers, 18(5) (1969), 401–409.
Somervuo, P.: Time topology for the self-organizing map. In Proceedings of IJCNN'99, volume 3, pages 1900–1905, Washington D.C., 1999.
TIMIT. DARPA TIMIT acoustic phonetic continuous speech database. CD-ROM, 1988.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Somervuo, P. Speech Dimensionality Analysis on Hypercubical Self-Organizing Maps. Neural Processing Letters 17, 125–136 (2003). https://doi.org/10.1023/A:1023646203167
Issue Date:
DOI: https://doi.org/10.1023/A:1023646203167