NeuralMUSIC: A Hybrid Neural-Subspace Framework for Robot Sound Source Localization
This work addresses the need for reliable sound source localization in robot audition, offering a hybrid approach that balances theoretical rigor and data-driven adaptability.
NeuralMUSIC combines a neural network for spatial covariance estimation with the classical MUSIC algorithm and a Frequency Attention Fusion module to improve robot sound source localization. It achieves competitive accuracy with enhanced robustness and cross-domain generalization.
Reliable sound source localization is fundamental to robot audition, enabling autonomous robots to perceive spatial cues and operate effectively in dynamic environments. Classical methods such as Multiple Signal Classification (MUSIC) offer strong theoretical foundations but degrade under low signal-to-noise ratios. While deep learning-based approaches achieve promising performance, they often struggle with limited generalization across conditions. To address these challenges, we propose NeuralMUSIC, a hybrid neural-subspace framework for robotic sound source localization. Specifically, a neural network first estimates the spatial covariance matrix from multichannel microphone observations. The predicted covariance is then integrated into a classical MUSIC pipeline with eigenvalue decomposition (EVD) and pseudo-spectrum computation, followed by a Frequency Attention Fusion (FAF) module to produce the final DOA estimates. To improve data efficiency, we further introduce a Self-supervised Spatial Correlation Learning (SSCL) strategy that leverages unlabeled acoustic data to capture spatial structure. Extensive experiments across different robotic tasks demonstrate that NeuralMUSIC achieves competitive localization accuracy while exhibiting improved robustness and cross-domain generalization.