Spatial Attention for Far-field Speech Recognition with Deep Beamforming Neural Networks

• Main Authors: He, Weipeng, Lu, Lu, Zhang, Biqiao, Mahadeokar, Jay, Kalgaonkar, Kaustubh, Fuegen, Christian
• Format: Journal Article
• Language: English
• Published: 05-11-2019
• Subjects: Computer Science - Sound
• DOI: 10.48550/arxiv.1911.02115

In this paper, we introduce spatial attention for refining the information in
multi-direction neural beamformer for far-field automatic speech recognition.
Previous approaches of neural beamformers with multiple look directions, such
as the factored complex linear projection, have shown promising results.
However, the features extracted by such methods contain redundant information,
as only the direction of the target speech is relevant. We propose using a
spatial attention subnet to weigh the features from different directions, so
that the subsequent acoustic model could focus on the most relevant features
for the speech recognition. Our experimental results show that spatial
attention achieves up to 9% relative word error rate improvement over methods
without the attention.