8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Variational Bayesian GMM for Speech Recognition

Fabio Valente, Christian Wellekens

Institut Eurecom, France

In this paper, we explore the potentialities of Variational Bayesian (VB) learning for speech recognition problems. VB methods deal in a more rigorous way with model selection and are a generalization of MAP learning. VB training for Gaussian Mixture Models is less affected than EM-ML training by over- fitting and singular solutions. We compare two types of Variational Bayesian Gaussian Mixture Models (VBGMM) with classical EM-ML GMM in a phoneme recognition task on the TIMIT database. VB learning performs better than EM-ML learning and is less affected by the initial model guess.

Full Paper

Bibliographic reference.  Valente, Fabio / Wellekens, Christian (2003): "Variational Bayesian GMM for speech recognition", In EUROSPEECH-2003, 441-444.