USE OF GENERALIZED DYNAMIC FEATURE PARAMETERS FOR SPEECH RECOGNITION

In this study, a new hidden Markov model that integrates generalized d ynamic feature parameters into the model structure is developed and ev aluated using maximum-likelihood (ML) and minimum-classification-error (MCE) pattern recognition approaches, in addition to the motivation o f direct minimization of error sate, the MCE approach automatically el iminates the necessity of artificial constraints, which were essential far the model formulation based on the ML approach, on the weighting functions in the definition of the generalized dynamic parameters, We design the loss function for minimizing error rate specifically for th e new model, and derive an analytical form of the gradient of the loss function that enables the implementation of the MCE approach, The con vergence property of the training procedure based on the MCE approach is investigated, and the experimental results from a standard TIMIT ph onetic classification task demonstrate a 13.4% error rate reduction co mpared with the ML approach.