8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Classification with Free Energy at Raised Temperatures

Rita Singh (1), Manfred K. Warmuth (2), Bhiksha Raj (3), Paul Lamere (4)

(1) Carnegie Mellon University, USA
(2) University of California at Santa Cruz, USA
(3) Mitsubishi Electric Research Laboratories, USA
(4) Sun Microsystems Laboratories, USA

In this paper we describe a generalized classification method for HMM-based speech recognition systems, that uses free energy as a discriminant function rather than conventional probabilities. The discriminant function incorporates a single adjustable temperature parameter T. The computation of free energy can be motivated using an entropy regularization, where the entropy grows monotonically with the temperature. In the resulting generalized classification scheme, the values of T = 0 and T = 1 give the conventional Viterbi and forward algorithms, respectively, as special cases. We show experimentally that if the test data are mismatched with the classifier, classification at temperatures higher than one can lead to significant improvements in recognition performance. The temperature parameter is far more effective in improving performance on mismatched data than a variance scaling factor, which is another apparent single adjustable parameter that has a very similar analytical form.

Full Paper

Bibliographic reference.  Singh, Rita / Warmuth, Manfred K. / Raj, Bhiksha / Lamere, Paul (2003): "Classification with free energy at raised temperatures", In EUROSPEECH-2003, 1773-1776.