ACL RD-TEC 1.0 Summarization of W03-0204
Paper Title:
PLASER: PRONUNCIATION LEARNING VIA AUTOMATIC SPEECH RECOGNITION
PLASER: PRONUNCIATION LEARNING VIA AUTOMATIC SPEECH RECOGNITION
Authors: Brian Mak and Manhung Siu and Mimi Ng and Yik-Cheung Tam and Yu-Chung Chan and Kin-Wah Chan and Ka-Yee Leung and Simon Ho and Jimmy Wong and Jacqueline Lo
Primarily assigned technology terms:
- acoustic modelling
- algorithm
- automatic speech recognition
- classification
- computer science
- computer-aided language learning
- computing
- discriminative training
- error reduction
- hidden markov
- hidden markov models
- hmms
- internet
- language learning
- learning
- matching
- modelling
- phoneme recognition
- phoneme recognizer
- phoneme verification
- pronunciation learning
- pronunciation scoring
- qualitative evaluation
- quantitative evaluation
- recognition
- recognition technology
- recognizer
- reporting
- scoring
- spectral subtraction
- speech recognition
- speech recognition technology
- spelling
- stochastic matching
- transcription
- two-class classification
- visualization
- viterbi
- weighting
Other assigned terms:
- accent
- acoustic models
- american english
- american english timit
- approach
- case
- chinese translation
- classification accuracy
- classification problem
- confidence score
- confidence scores
- data set
- dictionary
- duration
- evaluations
- fact
- feature
- foreign language
- gaussian mixture
- gaussian mixture model
- implementation
- index
- knowledge
- likelihood
- log-likelihood
- log-likelihood ratio
- markov models
- measure
- method
- model combination
- noise
- pause
- phoneme
- phonemes
- phonemic transcription
- posterior
- posterior probability
- probability
- pronunciation
- pronunciation quality
- questionnaire
- recognition accuracy
- relative error reduction
- segments
- sentence
- sentences
- sigmoid function
- silence
- speech data
- statistical models
- statistics
- symbols
- system design
- teaching
- technologies
- technology
- test set
- timit corpus
- training
- training data
- training set
- transcriptions
- user
- utterance
- viterbi path
- vocal tract
- vowel
- word
- words
