Cohen, M., Cohen, I., Rumelhart, D., Morgan, N., Franco, H., Abrash, V., & Konig, Y. (1992). Combining neural networks and hidden markov models for continuous speech recognition. In In ICSLP-92, 915—918.
We present a speaker-independent, continuous-speech recognition system based on a hybrid multilayer perceptron (MLP)/hidden Markov model (HMM). The system combines the advantages of both approaches by using MLPs to estimate the state-dependent observation probabilities of an HMM. New MLP architectures and training procedures are presented that allow the modeling of multiple distributions for phonetic classes and context-dependent phonetic classes. Comparisons with a pure HMM system illustrate advantages of the hybrid approach both in recognition accuracy and in number of parameters required.