Semi-supervised Speech Enhancement in Envelop and Details Subspaces
This work addresses speech intelligibility enhancement in noisy environments, representing an incremental improvement over existing methods.
The paper tackles speech enhancement by decoupling noisy speech spectrograms into spectral envelope and details subspaces, using supervised low-rank/sparse decomposition and Bayesian non-negative factorization to improve intelligibility, achieving satisfactory performance in perceptual quality and intelligibility compared to four existing algorithms.
In this study, we propose a modulation decoupling based single channel speech enhancement subspace framework, in which the spectrogram of noisy speech is decoupled as the product of a spectral envelop subspace and a spectral details subspace. This decoupling approach provides a method to specifically work on elimination of those noises that greatly affect the intelligibility. Two supervised low-rank and sparse decomposition schemes are developed in the spectral envelop subspace to obtain a robust recovery of speech components. A Bayesian formulation of non-negative factorization is used to learn the speech dictionary from the spectral envelop subspace of clean speech samples. In the spectral details subspace, a standard robust principal component analysis is implemented to extract the speech components. The validation results show that compared with four speech enhancement algorithms, including MMSE-SPP, NMF-RPCA, RPCA, and LARC, the proposed MS based algorithms achieve satisfactory performance on improving perceptual quality, and especially speech intelligibility.