Can I read Attention-based Neural Beamforming Layers for Multi-channel Speech Recognition on EtoBox?

Attention-based Neural Beamforming Layers for Multi-channel Speech Recognition by Pulugundla, Bhargav; Gao, Yang; King, Brian; Keskin, Gokce; Mallidi, Harish; Wu, Minhua; Droppo, Jasha; Maas, Roland is a scholarly article available to read on EtoBox.

What is Attention-based Neural Beamforming Layers for Multi-channel Speech Recognition about?

Attention-based beamformers have recently been shown to be effective for multi-channel speech recognition. However, they are less capable at capturing local information. In this work, we propose a 2D Conv-Attention module which combines convolution neural networks with attention for beamforming. We apply self- and cross-attention to explicitly model the correlations within and between the input channels. The end-to-end 2D Conv-Attention model is compared with a multi-head self-attention and superdirective-based neural beamformers. We train and evaluate on an in-house multi-channel dataset. The results show a relative improvement of 3.8% in WER by the proposed model over the baseline neural beamformer.

Author

Pulugundla, Bhargav; Gao, Yang; King, Brian; Keskin, Gokce; Mallidi, Harish; Wu, Minhua; Droppo, Jasha; Maas, Roland

Published

2021

Language

EN