Hybrid neural network/hidden Markov model continuous-speech recognition

n M In this paper we present a hybrid multilayer perceptron (MLP)/hidde arkov model (HMM) speaker-independent continuous-speech recogni-b tion system, in which the advantages of both approaches are combined y using MLPs to estimate the state-dependent observation probabilities p of an HMM. New MLP architectures and training procedures are resented which allow the modeling of multiple distributions for phonetic a p classes and context-dependent phonetic classes. Comparisons with ure HMM system illustrate advantages of the hybrid approach both in terms of recognition accuracy and number of parameters required.