A00-2023 & H adapted techniques used in speech recognition . Corpus-based statistical knowledge
A00-2014 parser compared to most current speech recognition systems . The CDG parser parses
A00-1005 advantage of the recent advances in speech recognition technology . Therefore , some
A00-1005 recognition system . In addition , the speech recognition system is far from perfect .
A00-1012 transcriptions produced by automatic speech recognition systems . An experiment which
A00-1029 Zadrozny et al , 1998 ) . The speech recognition grammars for these systems accepted
A00-2017 tasks is that of word selection in speech recognition ( Jelinek , 1998 ) . There ,
A00-1005 keywords and phrases into the speech recognition system . In addition , the speech
A00-1005 string . This is needed because the speech recognition system often breaks up the utterance
A00-1005 irrespective of the performance of the speech recognition engine for the user . In other
A00-1005 and phrases was provided to the speech recognition system as a text document . In
A00-1040 television and processed by automatic speech recognition ( ASR ) software . The participants
A00-1005 a user asked for service , the speech recognition system could not be grammar-based
A00-1029 counterexamples selected from speech recognition error logs or selects counter-examples
A00-2014 in Figure 1 that integrates a speech recognition component based on HMMs with
A00-2014 the accuracy of an integrated speech recognition and CDG parsing system . We compare
A00-1012 augmenting a standard trigram speech recognition model with information about
A00-1016 ; ( Martin et al. , 1998 ) ) . Speech recognition is performed using a version
A00-1044 word error rate 4.2 Automatic Speech Recognition ( ASR ) Figure 5-1 shows IdentiFinder
A00-1044 the errors made by the combined speech recognition and named entity system . Fourth
hide detail