Deep Feedforward Sequential Memory Network (DFSMN) has shown superior performance on speech recognition tasks. Based on this work, we propose a novel network architecture which introduces pyramidal memory structure to represent various context information in different layers. Additionally, res-CNN layers are added in the front to extract more sophisticated features as well. Together with lattice-free maximum mutual information (LF-MMI) and cross entropy (CE) joint training criteria, experimental results show that this approach achieves word error rates (WERs) of 3.62% and 10.89% respectively on Librispeech and LDC97S62 (Switchboard 300 hours) corpora. Furthermore, Recurrent neural network language model (RNNLM) rescoring is applied and a WER of 2.97% is obtained on Librispeech.
@article{arxiv.1810.11352,
title = {A novel pyramidal-FSMN architecture with lattice-free MMI for speech recognition},
author = {Xuerui Yang and Jiwei Li and Xi Zhou},
journal= {arXiv preprint arXiv:1810.11352},
year = {2018}
}