ACL RD-TEC 1.0 Summarization of H94-1092
Paper Title:
SEGMENT-BASED ACOUSTIC MODELS FOR CONTINUOUS SPEECH RECOGNITION
SEGMENT-BASED ACOUSTIC MODELS FOR CONTINUOUS SPEECH RECOGNITION
Authors: Mari Ostendorf and J. Robin Rohlicek
Primarily assigned technology terms:
- acoustic modeling
- algorithm
- caching
- classification
- continuous speech recognition
- distribution modeling
- fast search
- language modeling
- lattice search
- modeling
- multi-pass recognition
- multi-pass recognition scoring
- multi-pass search
- recognition
- recognition system
- resource management
- scoring
- search
- speaker-independent recognition
- speech recognition
Other assigned terms:
- acoustic models
- continuous speech
- correlation
- distribution
- duration
- duration model
- feature
- feature matrix
- feature vectors
- frame
- hierarchical model
- language model
- latex
- lattice
- mapping
- n-gram
- n-gram language model
- names
- phoneme
- recognition accuracy
- resource management task
- statistical models
- stochastic model
- theoretical framework
- training
- training data
- word
- word lattice
