ACL RD-TEC 1.0 Summarization of H92-1079
Paper Title:
LARGE VOCABULARY RECOGNITION OF WALL STREET JOURNAL SENTENCES AT DRAGON SYSTEMS
LARGE VOCABULARY RECOGNITION OF WALL STREET JOURNAL SENTENCES AT DRAGON SYSTEMS
Authors: James Baker and Janet Baker and Paul Bamberg and Kathleen Bishop and Larry Gillick and Vera Helman and Zezhen Huang and Yoshiko Ito and Stephen Lowe and Barbara Peskin and Robert Roth and Francesco Scattone
Primarily assigned technology terms:
- acoustic modeling
- algorithm
- baum-welch algorithm
- bayesian smoothing
- beam pruning
- clustering
- clustering algorithm
- continuous speech recognition
- continuous speech recognizer
- dynamic programming
- dynamic programming algorithm
- em algorithm
- estimator
- forward pass
- hmms
- interactive program
- k-means
- k-means clustering
- likelihood estimator
- maximum likelihood
- maximum likelihood estimator
- mixture modeling
- modeling
- optimization
- parameter optimization
- pic training
- processing
- programming algorithm
- pruning
- re-estimation
- recognition
- recognition algorithm
- recognition system
- recognizer
- resource management
- segmentation
- signal processing
- smoothing
- smoothing problem
- speech recognition
- speech recognition system
- speech recognizer
- training algorithm
- training procedure
- vocabulary recognition
Other assigned terms:
- acoustic models
- alphabet
- approach
- backoff
- beam
- benchmark
- case
- co-articulation
- context information
- continuous speech
- debugging
- distribution
- duration
- error rate
- fact
- frame
- hypotheses
- hypothesis
- implementation
- independence assumption
- labeling
- language model
- lexical entries
- likelihood
- markov models
- method
- mixture models
- nist
- phoneme
- phonemes
- priori
- probabilities
- probability
- probability density
- probability distribution
- probability distributions
- procedure
- process
- pronunciation
- punctuation
- real-time performance
- recognition task
- resource management task
- sentence
- sentences
- signal
- stress
- terms
- test data
- test set
- tokens
- training
- training data
- triphone
- vocabulary
- word
- word error rate
- word error rates
- words
