Von Mises-Fisher models in the total variability subspace for language recognition

Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. I. Lopez-Moreno, D. Ramos, J. Gonzalez-Dominguez, and J. Gonzalez-Rodriguez, "Von Mises-Fisher models in the total variability subspace for language recognition", IEEE Signal Processing Letters, vol. 18, no. 12, pp. 705-708, October 2011
This letter proposes a new modeling approach for the Total Variability subspace within a Language Recognition task. Motivated by previous works in directional statistics, von Mises-Fisher distributions are used for assigning language-conditioned probabilities to language data, assumed to be spherically distributed in this subspace. The two proposed methods use Kernel Density Functions or Finite Mixture Models of such distributions. Experiments conducted on NIST LRE 2009 show that the proposed techniques significantly outperform the baseline cosine distance approach in most of the considered experimental conditions, including different speech conditions, durations and the presence of unseen languages.
This work was supported by the Ministerio de Ciencia e Innovación under FPI Grant TEC2009-14719-C02-01 and cátedra UAM-Telefónica