Train composed Deep Denoising Autoencoder

Hello there,

Please I need help with composed of two DDAEs by MATLAB , where first and second DDAEs have different window lengths of one and three frames respectively. Noisy speech features are used as the input of the first DDAE and its output, along with one past and one future enhanced frames from outputs of the first DDAE, are given to the next DDAE whose window length would be three. The central frame of the Second DDAE will be considered as the target enhanced frame. Each DDAE hasThree hidden layers

Please Help