First European Conference on Speech Communication and Technology

Paris, France
September 27-29, 1989

Application of the Dempster-Shafer Theory of Evidence to Improved-Accuracy Isolated-Word Recognition

R. L. Bowles, Robert I. Damper

Department of Electronics and Computer Science, University of Southampton, Highfield, Southampton, UK

This paper describes experiments in which the outputs of three isolated-word recognition algorithms were combined to yield a lower average error rate than that achieved by any individual algorithm. The input tokens were simulated, fixed-length spectral speech patterns subjected to additive noise and either a "high" or "low" degree of time distortion. The three recognition techniques used were dynamic time warping, hidden Markov modelling and a multilayer perceptron. Two combination techniques were employed: the formula derived from the Dempster-Shafer (D-S) theory of evidence and simple majority voting (MV). D-S performed significantly better than MV under both time-distortion conditions. Evidence is also presented that the assumption of independent word scores, which is necessary for D-S theory to be strictly applicable, is questionable.

Full Paper

Bibliographic reference.  Bowles, R. L. / Damper, Robert I. (1989): "Application of the dempster-shafer theory of evidence to improved-accuracy isolated-word recognition", In EUROSPEECH-1989, 2384-2388.