ACL RD-TEC 1.0 Summarization of P02-1019
Paper Title:
PRONUNCIATION MODELING FOR IMPROVED SPELLING CORRECTION
PRONUNCIATION MODELING FOR IMPROVED SPELLING CORRECTION
Authors: Kristina Toutanova and Robert Moore
Primarily assigned technology terms:
- algorithm
- alignment algorithm
- automatic alignment
- channel modeling
- classification
- decision tree
- decision trees
- decomposition
- error reduction
- learning
- learning algorithm
- learning algorithms
- learning methods
- letter-to-phone conversion
- linear search
- machine learning
- machine learning algorithm
- machine learning algorithms
- machine learning methods
- matching
- modeling
- neural networks
- noisy channel model
- noisy channel modeling
- noisy channel spelling
- predictor
- pronunciation modeling
- reporting
- rescoring
- search
- spelling
- spelling correction
- splitting
- statistical spelling correction
- terminology
- training algorithm
- viterbi
- viterbi training
Other assigned terms:
- classification accuracy
- conditional probability
- context size
- data set
- dictionaries
- dictionary
- distribution
- error rate
- fact
- feature
- generation
- heuristic
- hypothesis
- implementation
- independence assumption
- interpolation
- language model
- meaning
- method
- n-gram
- n-gram model
- noisy channel
- phone-based model
- probabilities
- probability
- probability distribution
- pronunciation
- pronunciation dictionary
- sequence model
- source language
- sources of information
- statistics
- substring
- symbol
- term
- test set
- text
- training
- training data
- training samples
- training set
- tree
- trees
- trigram
- vowel
- window size
- word
- word error rate
- word frequencies
- word pronunciation
- words
