INTERSPEECH 2013
14thAnnual Conference of the International Speech Communication Association

Lyon, France
August 25-29, 2013

On Von-Mises Fisher Mixture Model in Text-Independent Speaker Identification

Jalil Taghia (1), Zhanyu Ma (2), Arne Leijon (1)

(1) KTH, Sweden
(2) BUPT, China

This paper addresses text-independent speaker identification (SI) based on line spectral frequencies (LSFs). The LSFs are transformed to differential LSFs (ĢLSF) in order to exploit their boundary and ordering properties. We show that the square root of ĢLSF has interesting directional characteristics implying that their distribution can be modeled by a mixture of von-Mises Fisher (vMF) distributions. We analytically estimate the mixture model parameters in a fully Bayesian treatment by using variational inference. In the Bayesian inference, we can potentially determine the model complexity and avoid overfitting problem associated with conventional approaches based on the expectation maximization. The experimental results confirm the effectiveness of the proposed SI system.

Full Paper

Bibliographic reference.  Taghia, Jalil / Ma, Zhanyu / Leijon, Arne (2013): "On von-mises fisher mixture model in text-independent speaker identification", In INTERSPEECH-2013, 2499-2503.