Advertisement

Summary

Chapter
  • 245 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2560)

Abstract

Automatic speech recognition (ASR) systems have reached a technological level that allows their usage in real applications. Leaning towards the human processing of speech various models need to be built to reflect these procedures. Since these models perform badly when there is a mismatch between the situation the system was faced with during the training stage of the models and the situation in which it is actually used or tested,adaptation methods are widely used to adapt the models to the current situation.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Personalised recommendations