Search

This paper proposes a new speech bandwidth expansion method, which uses Deep Neural Networks (DNNs) to build high-order eigenspaces between the low frequency components and the high frequency components of the speech signal. A four-layer DNN is trained layer-by-layer from a cascade of Neural Networks (NNs) and two Gaussian-Bernoulli Restricted Boltzmann Machines (GBRBMs). The GBRBMs are adopted to model the distribution of spectral envelopes of the low frequency and the high frequency respectively. The NNs are used to model the joint distribution of hidden variables extracted from the two GBRBMs. The proposed method takes advantage of the strong modeling ability of GBRBMs in modeling the distribution of the spectral envelopes. And both the objective and subjective test results show that the proposed method outperforms the conventional GMM based method.

In this paper, a data-driven temporal processing method based on Independent Component Analysis (ICA) is proposed for obtaining a more robust speech representation. Two different schemes of dominant temporal filters based on ICA are investigated. The one is the perceptuallybased filter which always focuses on the modulation frequency range between 1 and 16 Hz and the other is the most independent component discovered by ICA algorithm. Detailed comparative analysis between the proposed ICA-derived temporal filters and the previous statistical methods including Linear Discriminant Analysis (LDA) and Principle Component Analysis (PCA) is presented. The preliminary experiments show that the performance of the ICA based temporal filtering is much better in comparison with the LDA and PCA based methods in noisy environment.