Spatial Attention For Far-field Speech Recognition With Deep Beamforming Neural Networks

Abstract

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.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyhe2019spatial

Related papers

• A Unified Multichannel Far-field Speech Recognition System: Combining Neural Beamforming With Attention Based End-to-end Model (2024)0.00
• Enhanced Neural Beamformer With Spatial Information For Target Speech Extraction (2023)2.26
• Attention-based Neural Beamforming Layers For Multi-channel Speech Recognition (2021)0.00
• Locate And Beamform: Two-dimensional Locating All-neural Beamformer For Multi-channel Speech Separation (2023)3.58
• Dual-path Transformer Based Neural Beamformer For Target Speech Extraction (2023)0.00
• Deep Long Short-term Memory Adaptive Beamforming Networks For Multichannel Robust Speech Recognition (2017)13.23
• 3D Neural Beamforming For Multi-channel Speech Separation Against Location Uncertainty (2023)0.00
• Deep Ad-hoc Beamforming (2018)9.59