ACL RD-TEC 1.0 Summarization of H89-2045
Paper Title:
SPECTRAL ESTIMATION FOR NOISE ROBUST SPEECH RECOGNITION
SPECTRAL ESTIMATION FOR NOISE ROBUST SPEECH RECOGNITION
Authors: Adoram Erell and Mitch Weintraub
Primarily assigned technology terms:
- algorithm
- approximation
- computing
- continuous speech recognition
- database
- error estimation
- estimation algorithm
- estimation method
- estimator
- hmm recognition
- hmm recognition system
- matching
- modeling
- noise reduction
- noise reduction algorithm
- preprocessing
- processing
- recognition
- recognition system
- recognition system \
- recognition systems
- recognizer
- resource management
- robust speech recognition
- spectral estimation
- spectral subtraction
- speech recognition
- speech recognition system
- speech recognition systems
- statistical modeling
- template matching
- vector quantization
Other assigned terms:
- cepstral domain
- cepstral vector
- clean speech database
- continuous speech
- continuous-speech
- correlations
- distance metric
- distribution
- error rate
- estimation
- euclidean distance
- fact
- feature
- feature vector
- grammar
- implementation
- large training
- method
- nist
- noise
- noisy speech
- perplexity
- precision
- probabilities
- probability
- probability distribution
- probability distributions
- recognition accuracy
- recognition error rate
- recognition task
- resource management task
- sentence
- sentences
- signal
- speech data
- speech database
- statistical model
- term
- test set
- training
- training database
- training set
- vocabulary
- white noise
- word
