State of the art speech recognition systems can recognize continuously spoken speech from a large vocabulary in real time. In the future, portable speech recognition systems will be made possible by advances in Integrated circuit technology, by optlmizing system architectures, and by exploiting the special features of personal communications systems. This paper gives an overview of speech recognition systems and discusses design strategies for their use in portable communications.
T h i s paper presents t h e architecture of a w o r d processing subsystem for a large vocabulary real t i m e continuous speech recognition system. T h e syst e m contains three application specific integrated circuits to perform t h e Viterbi algorithm for 50,000 states using a breadth first recognition search i n real time. T h i s corresponds t o a c o m p u t a t i o n rate of 225 Megaoperations per second excluding m e m o r y operations where 670 Megabits have to b e accessed per second.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.