ACL RD-TEC 1.0 Summarization of P04-1030
Paper Title:
HEAD-DRIVEN PARSING FOR WORD LATTICES
HEAD-DRIVEN PARSING FOR WORD LATTICES
Authors: Christopher Collins and Bob Carpenter and Gerald Penn
Primarily assigned technology terms:
- algorithm
- beam search
- best-first search
- bottom-up chart parsing
- bottom-up parsing
- bottom-up parsing algorithm
- chart parsing
- cky algorithm
- cocke-kasami-younger parsing
- decoding
- dynamic programming
- expectation maximization
- head-driven lattice parsing
- head-driven parsing
- language modelling
- largevocabulary speech recognition
- lattice parsing
- list rescoring
- maximum likelihood
- measuring
- modelling
- n-best list parsing
- optimization
- parameter estimation
- parameterization
- parser
- parsing
- parsing algorithm
- pos tagging
- probabilistic parsing
- processor
- pruning
- random sampling
- recall degradation
- recognition
- recognizer
- rescoring
- right-branching
- sampling
- scoring
- search
- smoothing
- speech recognition
- speech recognizer
- speech understanding
- statistical parsing
- tagger
- tagging
- tokenization
- top-down parser
- transcription
- tuning
- word recognition
Other assigned terms:
- acoustic model
- acoustic score
- adjunction
- beam
- beam size
- bias
- complete parse
- conditional probabilities
- context free grammar
- convergence
- corpora
- correlation
- data set
- data structures
- edge weight
- entropy
- error rate
- estimation
- evaluation task
- evaluations
- generation
- gold standard
- grammar
- heuristics
- hypotheses
- hypothesis
- implementation
- knowledge
- language model
- language model quality
- language model score
- language models
- lattice
- lattice format
- lattices
- lexical coverage
- likelihood
- measure
- measures
- method
- model parameters
- n-best list
- n-gram
- n-gram language model
- n-gram model
- n-gram models
- nist
- oracle
- parameter space
- parameter values
- parse
- parse tree
- parsing accuracy
- parsing model
- parsing models
- parsing task
- part-of-speech
- partial parse
- pcfg
- penn treebank
- perplexity
- precision
- probabilities
- probability
- process
- punctuation
- recognition accuracy
- recognition task
- run-time
- search space
- sentence
- sentences
- speech recognition task
- statistical parsing model
- structural information
- structured language model
- syntactic information
- technique
- terminals
- terms
- test data
- test set
- text
- training
- training and test data
- training corpora
- training corpus
- training data
- training set
- training time
- transition probabilities
- tree
- treebank
- trees
- trigram
- trigram model
- understanding
- utterance
- vertex
- vocabulary
- wall street journal text
- word
- word error rate
- word error rates
- word lattice
- word lattices
- word lists
- word sequence
- words
