Learning Compact Recurrent Neural Networks
This work addresses memory and latency constraints for mobile deployment of speech recognition models, representing an incremental improvement in model compression.
The paper tackled the problem of deploying large recurrent neural networks (RNNs) on mobile devices by learning compact RNNs and LSTMs through low-rank factorizations and parameter sharing, achieving a 75% reduction in parameters with only a 0.3% increase in word error rate on a speech recognition task.
Recurrent neural networks (RNNs), including long short-term memory (LSTM) RNNs, have produced state-of-the-art results on a variety of speech recognition tasks. However, these models are often too large in size for deployment on mobile devices with memory and latency constraints. In this work, we study mechanisms for learning compact RNNs and LSTMs via low-rank factorizations and parameter sharing schemes. Our goal is to investigate redundancies in recurrent architectures where compression can be admitted without losing performance. A hybrid strategy of using structured matrices in the bottom layers and shared low-rank factors on the top layers is found to be particularly effective, reducing the parameters of a standard LSTM by 75%, at a small cost of 0.3% increase in WER, on a 2,000-hr English Voice Search task.