Title :
Stochastic modelling: From pattern classification to speech recognition and translation
Author_Institution :
Lehrstuhl fur Inf. VI, Tech. Hochschule Aachen, Germany
Abstract :
This paper gives an overview of the stochastic modelling approach in automatic speech recognition and language translation. Starting from the Bayes decision rule for minimum error rate, we present the stochastic modelling approach to speech recognition and analyze its characteristic properties. We discuss the advantages of stochastic modelling and extend it to the translation of written language.
Keywords :
Bayes methods; error analysis; language translation; pattern classification; speech recognition; stochastic systems; Bayes decision rule; language translation; minimum error rate; pattern classification; speech recognition; speech translation; stochastic modelling; stochastic modelling approach; written language translation; Automatic speech recognition; Computer science; Error analysis; Loudspeakers; Natural languages; Pattern classification; Speech processing; Speech recognition; Statistics; Stochastic processes;
Conference_Titel :
Pattern Recognition, 2000. Proceedings. 15th International Conference on
Print_ISBN :
0-7695-0750-6
DOI :
10.1109/ICPR.2000.903478
