English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper, we propose a multi-channel speech source separation with adeep neural network (DNN) which is trained under the condition that no cleansignal is available. As an alternative to a clean signal, the proposed methodadopts an estimated speech signal by an unsupervised speech source separationwith a statistical model. As a statistical model of microphone input signal, weadopts a time-varying spatial covariance matrix (SCM) model which includesreverberation and background noise submodels so as to achieve robustnessagainst reverberation and background noise. The DNN infers intermediatevariables which are needed for constructing the time-varying SCM. Speech sourceseparation is performed in a probabilistic manner so as to avoid overfitting toseparation error. Since there are multiple intermediate variables, a lossfunction which evaluates a single intermediate variable is not applicable.Instead, the proposed method adopts a loss function which evaluates the outputprobabilistic signal directly based on Kullback-Leibler Divergence (KLD).Gradient of the loss function can be back-propagated into the DNN through allthe intermediate variables. Experimental results under reverberant conditionsshow that the proposed method can train the DNN efficiently even when thenumber of training utterances is small, i.e., 1K.

Unsupervised Training for Deep Speech Source Separation with Kullback-Leibler Divergence Based Probabilistic Loss Function