Bidirectional LSTM Network with Ordered Neurons for Speech Enhancement

Xiaoqi Li, Yaxing Li, Yuanjie Dong, Shan Xu, Zhihui Zhang, Dan Wang, Shengwu Xiong

Speech enhancement aims to reduce the noise and improve the quality and intelligibility of noisy speech. Long short-term memory (LSTM) network frameworks have achieved great success on many speech enhancement applications. In this paper, the ordered neurons long short-term memory (ON-LSTM) network with a new inductive bias to differential the long/short-term information in each neuron is proposed for speech enhancement. Comparing the low-ranking neurons with short-term or local information, the high-ranking neurons which contain the long-term or global information always update less frequently for a wide range of influence. Thus, the ON-LSTM can automatically learn the clean speech information from noisy input and show better expressive ability. We also propose a rearrangement concatenation rule to connect the ON-LSTM outputs of forward and backward layers to construct the bidirectional ON-LSTM (Bi-ONLSTM) for further performance improvement. The experimental results reveal that the proposed ON-LSTM schemes produce better enhancement performance than the vanilla LSTM baseline. And visualization result shows that our proposed model can effectively capture clean speech components from noisy inputs.

 DOI: 10.21437/Interspeech.2020-2245

Cite as: Li, X., Li, Y., Dong, Y., Xu, S., Zhang, Z., Wang, D., Xiong, S. (2020) Bidirectional LSTM Network with Ordered Neurons for Speech Enhancement. Proc. Interspeech 2020, 2702-2706, DOI: 10.21437/Interspeech.2020-2245.

  author={Xiaoqi Li and Yaxing Li and Yuanjie Dong and Shan Xu and Zhihui Zhang and Dan Wang and Shengwu Xiong},
  title={{Bidirectional LSTM Network with Ordered Neurons for Speech Enhancement}},
  booktitle={Proc. Interspeech 2020},