Differentiable Artificial Reverberation

TL;DR

This paper introduces differentiable artificial reverberation models that enable end-to-end training for more accurate AR parameter estimation, overcoming computational bottlenecks by using FIR approximations of IIR filters.

Contribution

The paper proposes FIR-based approximations of IIR filters in DAR models, enabling efficient end-to-end training for AR parameter estimation in reverberation modeling.

Findings

End-to-end trained DAR models outperform non-end-to-end approaches.

FIR approximations significantly improve training speed on GPU.

Subjective listening tests favor the proposed method.

Abstract

Artificial reverberation (AR) models play a central role in various audio applications. Therefore, estimating the AR model parameters (ARPs) of a reference reverberation is a crucial task. Although a few recent deep-learning-based approaches have shown promising performance, their non-end-to-end training scheme prevents them from fully exploiting the potential of deep neural networks. This motivates the introduction of differentiable artificial reverberation (DAR) models, allowing loss gradients to be back-propagated end-to-end. However, implementing the AR models with their difference equations "as is" in the deep learning framework severely bottlenecks the training speed when executed with a parallel processor like GPU due to their infinite impulse response (IIR) components. We tackle this problem by replacing the IIR filters with finite impulse response (FIR) approximations with the frequency-sampling method. Using this technique, we implement three DAR models -- differentiable Filtered Velvet Noise (FVN), Advanced Filtered Velvet Noise (AFVN), and Delay Network (DN). For each AR model, we train its ARP estimation networks for analysis-synthesis (RIR-to-ARP) and blind estimation (reverberant-speech-to-ARP) task in an end-to-end manner with its DAR model counterpart. Experiment results show that the proposed method achieves consistent performance improvement over the non-end-to-end approaches in both objective metrics and subjective listening test results.