ACL RD-TEC 1.0 Summarization of H89-1005
Paper Title:
SPEAKER INDEPENDENT PHONETIC TRANSCRIPTION OF FLUENT SPEECH FOR LARGE VOCABULARY SPEECH RECOGNITION
SPEAKER INDEPENDENT PHONETIC TRANSCRIPTION OF FLUENT SPEECH FOR LARGE VOCABULARY SPEECH RECOGNITION
Authors: S. E. Levinson and M. Y. Liberman and A. Ljolje and L. G. Miller
Primarily assigned technology terms:
- algorithm
- computing
- continuous speech recognition
- database
- databases
- dp algorithm
- finite state
- hidden markov
- hidden markov model
- isolated word recognition
- large vocabulary continuous speech recognition
- large vocabulary speech recognition
- levenshtein
- lexical access
- markov model
- maximum entropy
- maximum likelihood
- optimization
- parameter estimation
- parameter optimization
- parameter reestimation
- parser
- parsing
- phonetic transcription
- reading
- recognition
- recognition system
- recognition systems
- recognizer
- reestimation
- resource management
- search
- speaker-dependent recognition
- speech recognition
- speech recognition system
- speech recognition systems
- transcription
- word recognition
Other assigned terms:
- acoustic signal
- alphabet
- approach
- bigram
- continuous speech
- derivational equivalence
- dictionary
- duration
- english dictionary
- entropy
- estimation
- experimental results
- finite set
- finite state grammar
- frame
- grammar
- large vocabulary speech
- lattice
- levenshtein distance
- lexicon
- likelihood
- markov chain
- measure
- phonetic alphabet
- procedure
- pronunciation
- recognition rate
- resource management task
- sentence
- sentences
- signal
- statistics
- symbols
- syntactic structure
- technique
- telephone network
- temporal structure
- timit database
- training
- training data
- training material
- transcriptions
- transition matrix
- vocabulary
- well-formed sentence
- word
- word lattice
