ACL RD-TEC 1.0 Summarization of J00-3003
Paper Title:
DIALOGUE ACT MODELING FOR AUTOMATIC TAGGING AND RECOGNITION OF CONVERSATIONAL SPEECH
DIALOGUE ACT MODELING FOR AUTOMATIC TAGGING AND RECOGNITION OF CONVERSATIONAL SPEECH
Authors: Andreas Stolcke and Noah Coccaro and Rebecca Bates and Paul Taylor and Carol Van Ess-Dykema and Klaus Ries and Elizabeth Shriberg and Daniel Jurafsky and Rachel Martin and Marie Meteer
Primarily assigned technology terms:
- 1-best recognition
- algorithm
- approximation
- automatic classification
- automatic learning
- automatic segmentation
- automatic speech recognition
- automatic tagging
- bayesian belief network
- classification
- classification approach
- classification trees
- classifier
- classifiers
- combined classifier
- computational linguistics
- computing
- continuous speech recognizer
- conversation analysis
- conversational speech recognition
- conversational system
- da classification
- da discrimination
- da mixture modeling
- da modeling
- da recognition
- database
- decision tree
- decision tree training
- decision trees
- decoding
- dialogue act classification
- dialogue modeling
- dialogue translation
- dialogue translation system
- dialogue understanding
- discourse modeling
- discourse processing
- discriminative training
- dynamic programming
- encoding
- event classification
- expectation-maximization
- finite-state automata
- finite-state grammar induction
- forward-backward algorithm
- frequency smoothing
- grammar adaptation
- grammar induction
- grouping
- hidden markov
- hidden markov model
- hidden markov modeling
- hidden markov models
- hmm induction
- hmms
- hypothesis scoring
- identification
- indexing
- induction
- induction algorithm
- labeler
- language modeling
- language processing
- lattice rescoring
- learning
- learning techniques
- likelihood estimate
- lvcsr
- machine learning
- markov model
- markov modeling
- maximum entropy
- maximum likelihood
- message classification
- mixture modeling
- mixture-of-posteriors approach
- model integration
- modeling
- modeling dialogue
- natural language processing
- neural network
- neural networks
- normalization
- off-line processing
- parsing
- probabilistic classifier
- probabilistic modeling
- processing
- prosodic classification
- prosodic modeling
- recognition
- recognizer
- regression
- rescoring
- scoring
- search
- search process
- segmentation
- semantic classification
- semantic classification trees
- shallow parsing
- smoothing
- speaker identification
- speech recognition
- speech recognizer
- speech technology
- statistical inference
- statistical language modeling
- statistical methods
- statistical modeling
- summarizer
- tag estimation
- tagging
- task-oriented dialogue
- transcription
- transformation-based learning
- translation system
- tree classifier
- tree-growing
- tuning
- user interface
- utterance segmentation
- vector quantization
- viterbi
- viterbi algorithm
- viterbi decoding
- weighting
- word insertion
- word recognition
- word-based classification
Other assigned terms:
- accent
- acoustic information
- acoustic likelihood
- acoustic model
- acoustic models
- acoustic score
- adjacency pair
- ambiguity
- annotation
- approach
- automata
- backoff
- bias
- bigram
- break
- cache
- case
- classification accuracy
- classification error
- classification model
- classification task
- classification tasks
- coherence
- community
- computational framework
- concept
- conditional probabilities
- context-free grammars
- continuous speech
- conversation
- conversational speech
- conversational telephone speech
- corpora
- correlation
- correlations
- cue words
- damsl
- data consortium
- decision tree model
- deduction
- dependency structure
- dialog
- dialog act
- dialogue act
- dialogue acts
- dialogue annotation
- dialogue grammar
- dialogue model
- dialogue structure
- dialogues
- discourse
- discourse coherence
- discourse context
- discourse grammar
- discourse model
- discourse models
- discourse structure
- discourse structure annotation
- discourse structures
- distribution
- duration
- empirical results
- english translations
- entropy
- entropy models
- error rate
- estimation
- fact
- feature
- feature space
- finite-state grammar
- genre
- grammar
- grammars
- human performance
- hypotheses
- hypothesis
- independence assumption
- information source
- interpolation
- intonation
- kappa
- kappa statistic
- knowledge
- labeling
- language model
- language models
- lattice
- lexical features
- lexical material
- likelihood
- linguistic
- linguistic data
- linguistic data consortium
- linguistics
- long-distance dependencies
- manual segmentation
- maps
- markov models
- markup
- maximum entropy models
- meaning
- measure
- measures
- message
- metaphor
- method
- methodology
- mixture models
- model parameters
- n-gram
- n-gram model
- n-gram models
- n-grams
- natural language
- noise
- noisy input
- opinion
- opinions
- oracle
- parameter values
- part-of-speech
- part-of-speech tags
- pauses
- perplexity
- perplexity reduction
- pitch
- pitch accent
- posterior
- posterior probability
- prior distribution
- prior probability
- probabilistic approach
- probabilistic framework
- probabilities
- probability
- probability model
- process
- prosodic feature
- prosodic features
- prosodic information
- prosody
- recognition accuracy
- recognition errors
- recognition task
- relation
- relative frequency
- rhetorical relations
- segmentation problem
- semantic
- shallow discourse structure
- signal
- signal-to-noise ratio
- speaker change
- speaker identity
- speaking rate
- speech act
- speech corpus
- speech input
- speech recognition errors
- speech recognition task
- speech signal
- spontaneous dialogue
- statistic
- statistical approach
- statistical language model
- statistical model
- statistical models
- statistical significance
- statistics
- stochastic context-free grammars
- subdialogue
- switchboard corpus
- syntax
- tag estimation accuracy
- tag set
- tags
- task-oriented corpora
- technologies
- technology
- terms
- test data
- test set
- textual input
- theory
- tokens
- tone
- topics
- topology
- training
- training corpus
- training data
- training set
- transcripts
- transition probabilities
- translations
- tree
- tree model
- trees
- trigram
- trigram language model
- trigram model
- turn-taking
- understanding
- unigram
- user
- utterance
- verb
- verbmobil corpus
- vocabulary
- wh-question
- word
- word boundaries
- word error rate
- word error rates
- word information
- word order
- word sequence
- word sequences
- word string
- word strings
- words
