POST: parallel object-oriented speech toolkit

The authors give a short overview of POST, a parallel speech toolkit that is distributed freeware to academic institutions. The underlying idea of POST is that large computational problems, like the ones involved in automatic speech recognition (ASR), can be solved more cost effectively by using the aggregate power and memory of many computers. In its current version (January 96) and amongst other things, POST can perform simple feature extraction, training and testing of word and subword hidden Markov models (HMMs) with discrete and multi-Gaussian statistical modelling. The implementation of the parallelism is discussed and an evaluation of the performance on a telephone database is presented. A short introduction to Parallel Virtual Machine (PVM), the library through which the parallelism is achieved, is also given.