Data-Driven Optimization Strategy of Microphone Array Configurations in Vehicle Environments

Microphone array speech enhancement is a crucial component of vehicle intelligence. However, the complex acoustic environments and the spatial constraints of array layouts present challenges for the design and implementation of microphone arrays in intelligent vehicles. This study proposes a data-driven optimization strategy for constructing the optimal microphone array configuration in-vehicle environments. We first developed a novel in-vehicle noise model that considers azimuth and elevation angles by defining a search region for microphone elements within a plane. Subsequently, based on the in-vehicle noise model, we conducted sound field modeling to ensure the designed microphone array is compatible with the complex acoustic environments inside vehicles. Utilizing this sound field model, we formulated a specialized optimization algorithm to devise the optimal configuration of the microphone array. Finally, the designed array configuration was constructed using a MEMS microphone array acquisition system, and the array performance was evaluated in real driving environments. Compared to conventional microphone array configurations, comprehensive experiments indicate that the designed microphone array enhances performance by increasing the STOI scores by 13.9%, improving the output SNR levels by 53.3%, and ensuring robustness in complex in-vehicle acoustic environments.

Recommended citation: @article{liu2024data, title={{Data-Driven Optimization Strategy of Microphone Array Configurations in Vehicle Environments}}, author={Liu, Lehai and Bi, Fengrong and Lin, Jiewei and Qi, Tongtong and Li, Xin}, journal={IEEE Transactions on Instrumentation and Measurement}, volume={73}, pages={6506310}, year={2024}, publisher={IEEE} }
Download Paper