We present a general approach to reduce the size of feed-forward deep neural
networks (DNNs). We propose a rank-constrained topology, which factors the weights
in the input layer of the DNN in terms of a low-rank representation: unlike
previous work, our technique is applied at the level of the filters learned at
individual hidden layer nodes, and exploits the natural two-dimensional
time-frequency structure in the input. These techniques are applied on a
small-footprint DNN-based keyword spotting task, where we find that we can reduce
model size by 75% relative to the baseline, without any loss in performance.
Furthermore, we find that the proposed approach is more effective at improving
model performance compared to other popular dimensionality reduction techniques,
when evaluated with a comparable number of parameters.