Robust speech recognition in reverberant environments using subband-based steady-state monaural and binaural suppression.
Hyung-Min ParkMatthew MaciejewskiChanwoo KimRichard M. SternPublished in: INTERSPEECH (2014)
Keyphrases
- steady state
- speech recognition
- subband
- speech signal
- noisy environments
- linear prediction
- filter bank
- wavelet transform
- markov chain
- image compression
- wavelet coefficients
- bit rate
- high frequency
- hidden markov models
- multiresolution
- wavelet decomposition
- language model
- frequency domain
- speaker identification
- feature vectors
- sound source
- pattern recognition
- wavelet domain
- speech recognition systems
- discrete wavelet transform
- multiscale
- queue length
- low frequency
- wavelet packet
- video coding
- computationally efficient
- spatial domain
- video quality
- bitstream
- image coding
- frequency band
- spectral analysis
- video signals
- feature extraction
- information retrieval