A continuous speech recognition system has to explore the search space with the only information given by the probabilities provided by a set of hidden Markov models representing the different sounds of the language and the restrictions imposed by the language model. The objective of this work is to include additional information, related to the estimated location of the boundary between two consecutive phonemes. This information is provided by an artificial recurrent neural network.
Bibliographic reference. Rubio, Antonio J. / Reilly, Ronan G. (1995): "Preliminary results on speech signal segmentation with recurrent neural networks", In EUROSPEECH-1995, 2197-2200.