ACL RD-TEC 1.0 Summarization of H92-1069
Paper Title:
SPEAKER-INDEPENDENT PHONE RECOGNITION USING BREF
SPEAKER-INDEPENDENT PHONE RECOGNITION USING BREF
Authors: Jean-Luc Gauvain and Lori F. Lamel
Primarily assigned technology terms:
- algorithm
- automatic segmentation
- bootstrap
- continuous speech recognition
- hidden markov
- hidden markov model
- hmm-based recognizer
- markov model
- phone recognition
- phonetic transcription
- reading
- recognition
- recognition systems
- recognizer
- segmentation
- speech recognition
- speech recognition systems
- transcription
- validation
- viterbi
- viterbi algorithm
- word recognition
Other assigned terms:
- baseline performance
- bigram
- continuous speech
- dictionary
- duration
- duration model
- evaluation data
- foreign words
- frame
- french
- generation
- grammar
- labeling
- language model
- large corpus
- manual segmentation
- names
- paragraphs
- pause
- performance evaluation
- phone language model
- phoneme
- phonemes
- prediction accuracy
- probabilities
- procedure
- pronunciation
- recognition accuracy
- segment boundaries
- segments
- sentences
- signal
- silence
- speech data
- speech signal
- statistics
- subcorpus
- test data
- test set
- text
- training
- training data
- transcriptions
- utterance
- vocabulary
- word
- words
