E91-1015 Representations coming from the speech understanding module contain recognition scores s
N04-1028 parser used in the natural language understanding module of the system is a robust , context-free
E09-1028 Our motivation to introduce the understanding module is to rescore the ASR output
H92-1015 performance of the Natural Language Understanding modules . This will provide an upper
A00-1010 hypotheses as input . It invokes the understanding module on a hypothesis to obtain a semantic
N09-2034 speech recognizers and speech understanding modules . ROVER ( Fiscus , 1997 ) tried
N04-1028 speech recognition and language understanding modules is discussed . In order to close
H92-1015 recognition and natural language understanding modules , we have developed techniques
P03-1070 parallel threads , allowing the Understanding Module ( UM ) to interpret the multiple
N04-1028 speech recognition and language understanding modules of the system when confronted
M91-1025 time to the ITP Natural Language Understanding Module . It prunes temporal and locative
N04-1028 be fully parsed by the language understanding module . In parallel , the confirmation
N09-2034 results can be used in the discourse understanding module and the dialogue management module
P01-1016 semantic representation . The Understanding Module then sends the output to Collagen
N12-1021 system base - line , the language understanding module was de - 3We did not take the
C88-1065 Winston \ -LSB- 1982 \ -RSB- ) . The understanding module analyzes English text and produces
N04-1028 recognized by the recognition and understanding modules ( Gustafson et al. , 1997 ) .
H94-1084 document application . The text understanding module is a generic , fast , robust
P06-2051 XML-databases . As the speech understanding module starts , it first reads these
P01-1012 ( ASR ) , the natural language understanding module ( NLU ) and the dialogue manager
hide detail