ACL RD-TEC 1.0 Summarization of H01-1034
Paper Title:
IMPROVING INFORMATION EXTRACTION BY MODELING ERRORS IN SPEECH RECOGNIZER OUTPUT
IMPROVING INFORMATION EXTRACTION BY MODELING ERRORS IN SPEECH RECOGNIZER OUTPUT
Authors: D.D. Palmer and M. Ostendorf
Primarily assigned technology terms:
- 1-best recognition
- algorithm
- asr system
- chunking
- collapsing
- confidence estimation
- confidence prediction
- coreference resolution
- decision tree
- decision trees
- decoding
- detection and tracking
- entity detection
- entity recognition
- error detection
- error detection and correction
- error handling
- estimator
- extraction system
- hmms
- identification
- ie system
- information extraction
- information extraction system
- information retrieval
- joint decoding
- language understanding
- lattice decoding
- linear interpolation
- modeling
- named entity detection
- named entity recognition
- ne recognition
- ne system
- parsing
- post-processing
- predictor
- processing
- re-estimation
- recognition
- recognizer
- search
- sentence chunking and parsing
- smoothing
- speech recognizer
- statistical approaches
- tdt evaluation
- transcription
- viterbi
- viterbi decoding
Other assigned terms:
- acoustic score
- annotator
- approach
- asr output
- asr transcript
- broadcast news
- broadcast news data
- conditional independence
- confidence score
- confidence scores
- correlation
- cross entropy
- decision tree model
- document
- duration
- entropy
- error rate
- estimation
- evaluation method
- evaluation test
- experimental results
- f-measure
- fact
- feature
- feature sets
- human annotator
- hypotheses
- hypothesis
- independence assumption
- information extraction task
- information states
- interpolation
- labeling
- language model
- language model score
- lattice
- linear model
- measures
- method
- model parameters
- named entities
- named entity
- names
- ne recognition performance
- opinions
- parse
- phrase
- phrase category
- posterior
- posterior probability
- precision
- probabilistic models
- probabilities
- probability
- process
- relative frequency
- search strategy
- sentence
- slot
- speech data
- statistical model
- technique
- term
- test data
- test set
- tokens
- topology
- training
- training and test data
- training corpus
- training data
- training set
- transcript
- transcriptions
- tree
- tree model
- trees
- understanding
- utterance
- vocabulary
- vocabulary size
- word
- word duration
- word error rate
- word sequence
- words
