An invertible frequency eigendomain transformation for masking-based subspace speech enhancement.
Chang Huai YouSoo Ngee KohSusanto RahardjaPublished in: IEEE Signal Process. Lett. (2005)
Keyphrases
- speech enhancement
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- speech signal
- linear prediction
- wiener filter
- high dimensional
- feature space
- additive noise
- vocal tract
- feature extraction
- principal component analysis
- smoothing algorithm
- image restoration
- human visual system
- image quality
- markov random field
- background noise