ACL RD-TEC 1.0 Summarization of H89-2038
Paper Title:
LARGE-VOCABULARY SPEAKER-INDEPENDENT CONTINUOUS SPEECH RECOGNITION WITH SEMI.CONTINUOUS HIDDEN MARKOV MODELS
LARGE-VOCABULARY SPEAKER-INDEPENDENT CONTINUOUS SPEECH RECOGNITION WITH SEMI.CONTINUOUS HIDDEN MARKOV MODELS
Authors: X.D. Huang and H.W. Hon and K.F. Lee
Primarily assigned technology terms:
- algorithm
- classification
- clustering
- clustering algorithm
- computing
- continuous speech recognition
- database
- em algorithm
- error reduction
- estimation algorithm
- forward-backward algorithm
- hidden markov
- hidden markov model
- hidden markov modeling
- hidden markov models
- hmm re-estimation
- hmm-based speech recognition
- hmms
- independent continuous speech recognition
- k-means
- k-means clustering
- likelihood estimation
- likelihood maximization
- markov model
- markov modeling
- maximum likelihood
- maximum likelihood estimation
- modeling
- parameter estimation
- processing
- re-estimation
- re-estimation algorithm
- recognition
- recognition system
- resource management
- schmm re-estimation
- smoothing
- speaker-independent continuous speech recognition
- speech recognition
- speech recognition system
- t000-word speaker-independent continuous speech recognition
- vector quantization
- viterbi
- viterbi alignment
- weighting
Other assigned terms:
- approach
- codebook
- coefficient
- computational complexity
- continuous speech
- correlation
- correlations
- covariance matrix ~
- density function
- distribution
- distribution matrix
- error rate
- estimation
- events
- experimental results
- feature
- frame
- grammar
- hmm parameter
- implementation
- index
- likelihood
- linear frequency
- markov chain
- markov models
- methodology
- model parameters
- modeling power
- perplexity
- probabilistic framework
- probabilities
- probability
- probability density
- probability density function
- probability distribution
- probability distribution matrix
- procedure
- projection
- recognition accuracy
- recognition grammar
- recognition task
- segments
- sentences
- speaker-independent continuous speech recognition task
- speech recognition task
- speech vector
- symbols
- technique
- training
- training data
- transformation
- transition matrix
- transition probability
- triphone
- vocabulary
- word
- word-pair grammar
- words
