Replacing uncertainty decoding with subband re-estimation for large vocabulary speech recognition in noise.
Jianhua LuJi MingRoger F. WoodsPublished in: INTERSPEECH (2009)
Keyphrases
- speech recognition
- subband
- noisy environments
- low frequency
- filter bank
- wavelet transform
- image compression
- high frequency
- multiresolution
- bit rate
- wavelet coefficients
- speech recognizer
- frequency domain
- feature vectors
- hidden markov models
- speech recognition systems
- language model
- speech signal
- wavelet decomposition
- automatic speech recognition
- subband coding
- wavelet domain
- subband decomposition
- pattern recognition
- additive noise
- speaker identification
- speaker independent
- speaker adaptation
- discrete wavelet transform
- wavelet packet
- frequency band
- image coding
- computer vision
- noise model
- image coder
- signal to noise ratio
- vector quantization
- non stationary
- noise level
- transform coefficients
- video signals
- image segmentation
- neural network