Efficient and Microphone-Fault-Tolerant 3D Sound Source Localization

2025-05-27

AI Summary

To address key bottlenecks in three-dimensional sound source localization under complex environments—namely, high computational overhead, strong dependence on microphone geometry, and poor fault tolerance—this paper proposes a robust and efficient method. Methodologically, it introduces: (1) a novel sparse cross-attention mechanism to reduce computational redundancy; (2) an adaptive signal coherence metric coupled with microphone-position-agnostic representation learning, significantly enhancing robustness against unknown or failed microphones; and (3) a lightweight pretraining architecture enabling few-shot deployment. Experimental results demonstrate centimeter-level localization accuracy in multi-source scenarios, a 62% reduction in computational cost, stable operation with only four microphones, and sustained success rates exceeding 90% even when two microphones fail.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Hardware-aware ML
• Natural Language Processing: Safety and Robustness

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Sound source localization (SSL) is a critical technology for determining the position of sound sources in complex environments. However, existing methods face challenges such as high computational costs and precise calibration requirements, limiting their deployment in dynamic or resource-constrained environments. This paper introduces a novel 3D SSL framework, which uses sparse cross-attention, pretraining, and adaptive signal coherence metrics, to achieve accurate and computationally efficient localization with fewer input microphones. The framework is also fault-tolerant to unreliable or even unknown microphone position inputs, ensuring its applicability in real-world scenarios. Preliminary experiments demonstrate its scalability for multi-source localization without requiring additional hardware. This work advances SSL by balancing the model's performance and efficiency and improving its robustness for real-world scenarios.

Problem

• Achieve accurate 3D sound localization with low computational cost
• Enable fault-tolerant localization with unreliable microphone inputs
• Scale multi-source localization without additional hardware requirements

Innovation

• Uses sparse cross-attention for efficient processing
• Employs adaptive signal coherence metrics
• Fault-tolerant to unreliable microphone inputs