Applying Automatic Differentiation to Optimize Differential Microphone Array Designs

TL;DR

This paper presents a differentiable programming approach to optimize the design of differential microphone arrays, achieving efficient, cost-effective, and wide-band sound capture with adaptive constraints.

Contribution

It introduces a novel differentiable convex optimization method for designing adaptive microphone arrays with constrained placement and performance.

Findings

Achieves desired directivity over wide frequency range

Reduces implementation costs for wide-band speech recovery

Enables adaptive, constrained array design using automatic differentiation

Abstract

This paper introduces a novel methodology leveraging differentiable programming to design efficient, constrained adaptive non-uniform Linear Differential Microphone Arrays (LDMAs) with reduced implementation costs. Utilizing an automatic differentiation framework, we propose a differentiable convex approach that enables the adaptive design of a filter with a distortionless constraint in the desired sound direction, while also imposing constraints on microphone positioning to ensure consistent performance. This approach achieves the desired Directivity Factor (DF) over a wide frequency range and facilitates effective recovery of wide-band speech signals at lower implementation costs.