ACL RD-TEC 1.0 Summarization of P00-1031
Paper Title:
A NEW STATISTICAL APPROACH TO CHINESE PINYIN INPUT
A NEW STATISTICAL APPROACH TO CHINESE PINYIN INPUT
Authors: Zheng Chen and Kai-Fu Lee
Primarily assigned technology terms:
- beam search
- error reduction
- hidden markov
- hidden markov model
- hmms
- language modelling
- markov model
- modelling
- phonetic spelling
- pinyin-to-character conversion
- recognition
- search
- segmentation
- sentence-based input method
- speech recognition
- spelling
- spelling correction
- statistical language modelling
- trigram-based statistical language modelling
- viterbi
- viterbi beam
- viterbi beam search
Other assigned terms:
- acoustic model
- acronym
- approach
- beam
- bigram
- character sequence
- characters
- chinese characters
- chinese language
- chinese language model
- chinese word
- chinese words
- concept
- distribution
- document
- english spelling model
- error rate
- feature
- geometric mean
- language model
- large training
- lexicon
- mandarin chinese
- mapping
- markov models
- measure
- measures
- method
- model parameters
- n-gram
- perplexity
- pinyin
- pinyin input
- priori
- probabilities
- probability
- pronunciation
- proper name
- spelling model
- statistical approach
- statistical data
- statistical language model
- syllables
- system performance
- technology
- testing corpus
- testing data
- text
- theory
- training
- training corpus
- training data
- transition probability
- transposition
- trigram
- trigram language model
- trigram model
- unigram
- unigram language model
- user
- word
- word perplexity
- word sequence
- word sequences
- word string
- words
