# A novel pyramidal-FSMN architecture with lattice-free MMI for speech recognition

arXiv: https://arxiv.org/abs/1810.11352 · HTML: https://mlsanity.com/paper/1810.11352/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Oct 2018.

## Built on by (most cited first, 3 of 3)

- [BART based semantic correction for Mandarin automatic speech recognition system](https://mlsanity.com/paper/2104.05507/) (2021)
- [BiFSMN: Binary Neural Network for Keyword Spotting](https://mlsanity.com/paper/2202.06483/) (2022)
- [BiFSMNv2: Pushing Binary Neural Networks for Keyword Spotting to Real-Network Performance](https://mlsanity.com/paper/2211.06987/) (2022)

## Abstract (arXiv)

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.