Using STFT real and imaginary parts of modulation signals for MMSE-based speech enhancement.
Belinda SchwerinKuldip K. PaliwalPublished in: Speech Commun. (2014)
Keyphrases
- speech enhancement
- single channel
- real and imaginary parts
- frequency domain
- multi channel
- mexican hat
- noisy environments
- short time fourier transform
- independent component analysis
- noise reduction
- wiener filter
- gabor filters
- signal processing
- spatial domain
- wavelet decomposition
- prior information
- brain activity
- heart rate
- wavelet analysis
- linear prediction
- wavelet function
- signal to noise ratio
- feature extraction
- speech signal
- similarity measure
- multiscale
- probabilistic model
- wavelet basis
- background noise
- sound source
- low frequency
- activation function
- computer vision