Word recognition and incremental learning based on neural associative memories and hidden Markov models

An architecture for achieving word recognition and incremen- tal learning of new words in a language processing system is presented. The architecture is based on neural associative memories and hidden Markov models. The hidden Markov models generate subword-unit transcriptions of the spoken words and provide them as input to the associative mem- ory module. The associative memory module is a network of binary auto- and heteroassociative memories and responsible for combining words from subword-units. The basic version of the system is implemented for simple command sentences. Its performance is compared with the performance of the hidden Markov models.