Fast match acoustic models in large vocabulary continuous speech recognition

First pass acoustic models in a two pass, large vocabulary, speaker dependent, continuous speech recognizer are discussed. The models are based on single state VQ, compared to our earlier 5-state continuous HMMs. The loss in performance was recovered using context dependent models, multiple codebooks and improvements in the computation of the observation vector. This resulted in a first pass inclusion of >97% while reducing the computational burden by a factor of 30. Currently, the system runs real-time on an AT&T DSP32C module and an HP-735 workstation, with recognition accuracies of up to 90% for a 5000 word vocabulary task