I want to make Denoise Speech Using Deep Learning Networks example using LPS features?

I created logarithmic power spectra of the sounds. How do I create a network using LPS in this example?

cleanLPS = 10*log(cleanSTFT.^2);
noisyLPS = 10*log(noisySTFT.^2);

noisySTFT = [noisySTFT(:,1:numSegments - 1), noisySTFT];
stftSegments = zeros(numFeatures, numSegments , size(noisySTFT,2) - numSegments + 1);
for index = 1:size(noisySTFT,2) - numSegments + 1
    stftSegments(:,:,index) = (noisySTFT(:,index:index + numSegments - 1)); 
end
%% targets and predictors
targets = cleanSTFT;
size(targets)

predictors = stftSegments;
size(predictors)