Improved noise-robustness in distributed speech recognition via perceptually-weighted vector quantisation of filterbank energies.
Stephen SoKuldip K. PaliwalPublished in: INTERSPEECH (2005)
Keyphrases
- speech recognition
- filter bank
- noisy environments
- vector quantisation
- hidden markov models
- speech signal
- language model
- pattern recognition
- subband
- speech synthesis
- automatic speech recognition
- speech recognizer
- image coding
- spectral analysis
- speaker identification
- multiresolution
- background noise
- multiscale
- speech recognition systems
- frequency domain
- machine learning
- noise reduction
- image processing
- signal processing
- human visual system
- computationally efficient
- probabilistic model