Speech enhancement based on a log-spectral amplitude estimator and a postfilter derived from clean speech codebook.
Jason WungBiing-Hwang Fred JuangBowon LeePublished in: EUSIPCO (2010)
Keyphrases
- speech enhancement
- noisy speech
- linear prediction
- speech signal
- noisy environments
- speech recognition
- noise reduction
- signal to noise ratio
- spectral subtraction
- single channel
- hidden markov models
- vocal tract
- background noise
- vector quantization
- spectral analysis
- filter bank
- image coding
- subband
- speaker identification
- prediction error
- automatic speech recognition
- multi channel
- least squares
- wiener filter
- smoothing algorithm
- feature vectors
- speech synthesis
- edge detection
- multiresolution
- additive noise
- image representation
- image compression
- principal component analysis