Continuous speech recognition using an associative string processor

A description is given of research in the use of parallel computer architectures for continuous word speech recognition. In particular, the implementation of a continuous speech recognition algorithm on the associative string processor (ASP) is reported. The general principle of connected-word recognition, based on the Viterbi recognition algorithm, is given, followed by a brief description of the ASP architecture. The implementation on the ASP of an adaptation of the above connected-word recognition algorithm is described. The implementation exploits the parallelism inherent in the recognition algorithm and is simplified by the associative nature of the ASP architecture. The suitability of the ASP architecture for such speed recognition application is appraised