We propose a fully convolutional sequence-to-sequence encoder architecture with a simple and efficient decoder. Our model improves WER on LibriSpeech while being an order of magnitude more efficient than a strong RNN baseline. Key to our approach is a time-depth separable convolution block which dramatically reduces the number of parameters in the model while keeping the receptive field large. We also give a stable and efficient beam search inference procedure which allows us to effectively integrate a language model. Coupled with a convolutional language model, our time-depth separable convolution architecture improves by more than 22% relative WER over the best previously reported sequence-to-sequence results on the noisy LibriSpeech test set.
@article{arxiv.1904.02619,
title = {Sequence-to-Sequence Speech Recognition with Time-Depth Separable Convolutions},
author = {Awni Hannun and Ann Lee and Qiantong Xu and Ronan Collobert},
journal= {arXiv preprint arXiv:1904.02619},
year = {2019}
}