Differentiable Attenuation Filters for Feedback Delay Networks

2025-11-25

AI Summary

This work addresses the challenge of designing attenuation filters for Feedback Delay Networks (FDNs) in digital audio reverberation. We propose a lightweight, differentiable parametrization scheme based on second-order IIR filters configured as frequency-selective equalizers. To reduce parameter count while preserving full differentiability, we introduce cross-delay-line sharing of center frequencies and Q-factors, adapting only the gain per delay length. The design integrates analog filter principles with supervised learning, supports flexible configuration of filter count, and enables end-to-end gradient-based optimization. Experiments demonstrate that our method achieves state-of-the-art accuracy in frequency-dependent decay control and high-fidelity reverberation timbre, while drastically reducing computational overhead. It offers superior flexibility, strong scalability, and training efficiency compared to prior approaches.

Technology Category

• Natural Language Processing: Learning & Optimization for NLP
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

We introduce a novel method for designing attenuation filters in digital audio reverberation systems based on Feedback Delay Net- works (FDNs). Our approach uses Second Order Sections (SOS) of Infinite Impulse Response (IIR) filters arranged as parametric equalizers (PEQ), enabling fine control over frequency-dependent reverberation decay. Unlike traditional graphic equalizer designs, which require numerous filters per delay line, we propose a scal- able solution where the number of filters can be adjusted. The fre- quency, gain, and quality factor (Q) parameters are shared parame- ters across delay lines and only the gain is adjusted based on delay length. This design not only reduces the number of optimization parameters, but also remains fully differentiable and compatible with gradient-based learning frameworks. Leveraging principles of analog filter design, our method allows for efficient and accu- rate filter fitting using supervised learning. Our method delivers a flexible and differentiable design, achieving state-of-the-art per- formance while significantly reducing computational cost.

Problem

• Designing scalable attenuation filters for audio reverberation systems
• Reducing optimization parameters while maintaining differentiability
• Achieving efficient frequency-dependent decay control with gradient-based learning

Innovation

• Attenuation filters using Second Order Sections
• Shared parameters across delay lines
• Differentiable design for gradient-based learning