Multiresolution channel normalization for ASR in reverberant environments.
Carlos AvendañoSangita TibrewalaHynek HermanskyPublished in: EUROSPEECH (1997)
Keyphrases
- multiresolution
- speech signal
- automatic speech recognition
- preprocessing
- wavelet transform
- quadtree
- dynamic environments
- speech recognition
- wavelet decomposition
- multi channel
- variable resolution
- multiple access
- multiresolution representation
- wavelet coefficients
- noisy environments
- real world
- normalization method
- image fusion
- noisy channel
- highly dynamic
- coarse to fine
- subband
- multiscale
- image processing