5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Information Theoretic Approaches to Model Selection

Jonathan Hamaker, Aravind Ganapathiraju, Joseph Picone

Institute for Signal and Information Processing, Mississippi State University, USA

The primary problem in large vocabulary conversational speech recognition (LVCSR) is poor acoustic-level matching due to large variability in pronunciations. There is much to explore about the "quality" of states in an HMM and the inter-relationships between inter-state and intra-state Gaussians used to model speech. Of particular interest is the variable discriminating power of the individual states. The fundamental concept addressed in this paper is to investigate means of exploiting such dependencies through model topology optimization based on the Bayesian Information Criterion (BIC) and the Minimum Description Length (MDL) principle.

Full Paper

Bibliographic reference.  Hamaker, Jonathan / Ganapathiraju, Aravind / Picone, Joseph (1998): "Information theoretic approaches to model selection", In ICSLP-1998, paper 0653.