#19977Though some details of the CAS are particular to individual domains, the Comparator software is domain-independent, as is the CAS approach.Spoken Language Systems will require integration of continuous speech recognition and natural language processing.
tech,18-2-H89-2028,ak
speech recognizer ( CSR )
</term>
and a
<term>
natural language processor ( NLP )
</term>
to form a
<term>
spoken language system
#20010This is a proposed specification for an interface between a continuous speech recognizer (CSR) and anatural language processor ( NLP ) to form a spoken language system.
tech,10-2-H89-2028,ak
an
<term>
interface
</term>
between a
<term>
continuous speech recognizer ( CSR )
</term>
and a
<term>
natural language processor
#20002This is a proposed specification for an interface between acontinuous speech recognizer ( CSR ) and a natural language processor (NLP) to form a spoken language system.
tech,5-4-H89-2028,ak
specification
</term>
also defines a
<term>
Top-N mode
</term>
in which a first part outputs a list
#20043The specification also defines aTop-N mode in which a first part outputs a list of top N scored sentences for postprocessing by a second part.
tech,27-2-H89-2028,ak
processor ( NLP )
</term>
to form a
<term>
spoken language system
</term>
. Both components are integrated
#20019This is a proposed specification for an interface between a continuous speech recognizer (CSR) and a natural language processor (NLP) to form aspoken language system.
tech,6-3-H89-2028,ak
Both components are integrated with a
<term>
stack controller
</term>
and contribute to the
<term>
search
#20029Both components are integrated with astack controller and contribute to the search control.
tech,7-2-H89-2028,ak
proposed
<term>
specification
</term>
for an
<term>
interface
</term>
between a
<term>
continuous speech
#19999This is a proposed specification for aninterface between a continuous speech recognizer (CSR) and a natural language processor (NLP) to form a spoken language system.
tech,11-1-H89-2028,ak
continuous speech recognition
</term>
and
<term>
natural language processing
</term>
. This is a proposed
<term>
specification
#19988Spoken Language Systems will require integration of continuous speech recognition andnatural language processing.
tech,8-5-H89-2028,ak
<term>
specification
</term>
might be
<term>
NLP evaluation testing
</term>
: a common simulated
<term>
CSR
</term>
#20073An additional use for this specification might beNLP evaluation testing: a common simulated CSR could be interfaced to each site's NLP to provide identical testing environments.
tech,21-4-H89-2028,ak
list of top N scored sentences for
<term>
postprocessing
</term>
by a second part . An additional
#20059The specification also defines a Top-N mode in which a first part outputs a list of top N scored sentences forpostprocessing by a second part.
tech,27-5-H89-2028,ak
<term>
NLP
</term>
to provide identical
<term>
testing environments
</term>
. A model is presented to characterize
#20092An additional use for this specification might be NLP evaluation testing: a common simulated CSR could be interfaced to each site's NLP to provide identicaltesting environments.
tech,7-1-H89-2028,ak
</term>
will require integration of
<term>
continuous speech recognition
</term>
and
<term>
natural language processing
#19984Spoken Language Systems will require integration ofcontinuous speech recognition and natural language processing.
model,4-2-H89-2028,ak
processing
</term>
. This is a proposed
<term>
specification
</term>
for an
<term>
interface
</term>
between
#19996This is a proposedspecification for an interface between a continuous speech recognizer (CSR) and a natural language processor (NLP) to form a spoken language system.
tech,23-5-H89-2028,ak
could be interfaced to each site 's
<term>
NLP
</term>
to provide identical
<term>
testing
#20088An additional use for this specification might be NLP evaluation testing: a common simulated CSR could be interfaced to each site'sNLP to provide identical testing environments.
tech,15-5-H89-2028,ak
testing
</term>
: a common simulated
<term>
CSR
</term>
could be interfaced to each site
#20080An additional use for this specification might be NLP evaluation testing: a common simulatedCSR could be interfaced to each site's NLP to provide identical testing environments.
model,1-4-H89-2028,ak
the
<term>
search control
</term>
. The
<term>
specification
</term>
also defines a
<term>
Top-N mode
</term>
#20039Thespecification also defines a Top-N mode in which a first part outputs a list of top N scored sentences for postprocessing by a second part.
tech,12-3-H89-2028,ak
controller
</term>
and contribute to the
<term>
search control
</term>
. The
<term>
specification
</term>
also
#20035Both components are integrated with a stack controller and contribute to thesearch control.
model,5-5-H89-2028,ak
part . An additional use for this
<term>
specification
</term>
might be
<term>
NLP evaluation testing
#20070An additional use for thisspecification might be NLP evaluation testing: a common simulated CSR could be interfaced to each site's NLP to provide identical testing environments.