ACL RD-TEC 1.0 Summarization of H94-1014
Paper Title:
LANGUAGE MODELING WITH SENTENCE-LEVEL MIXTURES
LANGUAGE MODELING WITH SENTENCE-LEVEL MIXTURES
Authors: Rukmini Iyer and Mari Ostendorf and J. Robin Rohlicek
Primarily assigned technology terms:
- agglomerative clustering
- algorithm
- atis
- automatic learning
- automatic topic clustering
- clustering
- clustering algorithm
- computing
- continuous speech recognizer
- decision tree
- dynamic language modeling
- em algorithm
- error correction
- estimation algorithm
- expectation-maximization
- grouping
- hidden markov
- hidden markov model
- hmms
- identification
- iterative algorithm
- language model adaptation
- language model training
- language modeling
- learning
- learning techniques
- markov model
- maximum likelihood
- mixture weight estimation
- model adaptation
- model training
- modeling
- n-best reranking
- n-best rescoring
- normalization
- parameter estimation
- partitioning
- ranking
- re-estimation
- recognition
- recognition search
- recognition system
- recognizer
- reranking
- rescoring
- search
- smoothing
- speech recognizer
- speech transcription
- topic clustering
- topic initialization
- topic spotting
- transcription
- tree clustering
- weight estimation
- witten-bell back-off
Other assigned terms:
- acoustic model
- approach
- benchmark
- bigram
- byblos system
- cache
- case
- cluster
- clusters
- content words
- context-free grammar
- context-free grammars
- continuous speech
- data set
- dialog
- distribution
- document
- dynamic language model
- dynamic model
- error rate
- estimation
- evaluation test
- experimental results
- formalism
- function words
- grammar
- grammars
- heuristic
- heuristic rule
- heuristic rules
- hypotheses
- hypothesis
- implementation
- index
- interpolation
- knowledge
- language model
- language models
- likelihood
- long distance dependencies
- long-distance dependencies
- measure
- measures
- mechanisms
- method
- mixture models
- multinomial distribution
- n-gram
- n-gram model
- n-gram models
- n-grams
- normalization factor
- paragraph
- paragraphs
- part-of-speech
- perplexity
- priori
- probabilities
- probability
- probability estimates
- process
- pronunciation
- punctuation
- recognition accuracy
- recognition errors
- search strategy
- sentence
- sentence level
- sentences
- sequence model
- similarity measures
- speaking style
- statistical language model
- statistics
- stochastic language model
- stochastic segment model
- style
- system performance
- technique
- term
- terms
- test set
- text
- text corpus
- topics
- training
- training data
- transcriptions
- tree
- trigram
- trigram language model
- utterance
- verb
- vocabulary
- word
- word count
- word error rate
- word sequence
- word sequences
- word string
- words
