# Lightweight and interpretable neural modeling of an audio distortion effect using hyperconditioned differentiable biquads

arXiv: https://arxiv.org/abs/2103.08709 · HTML: https://mlsanity.com/paper/2103.08709/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Mar 2021.

## Built on by (most cited first, 10 of 10)

- [Modelling black-box audio effects with time-varying feature modulation](https://mlsanity.com/paper/2211.00497/) (2022)
- [Direct design of biquad filter cascades with deep learning by sampling random polynomials](https://mlsanity.com/paper/2110.03691/) (2021)
- [Differentiable All-pole Filters for Time-varying Audio Systems](https://mlsanity.com/paper/2404.07970/) (2024)
- [Rigid-Body Sound Synthesis with Differentiable Modal Resonators](https://mlsanity.com/paper/2210.15306/) (2022)
- [Differentiable Grey-box Modelling of Phaser Effects using Frame-based Spectral Processing](https://mlsanity.com/paper/2306.01332/) (2023)
- [Unsupervised Estimation of Nonlinear Audio Effects: Comparing Diffusion-Based and Adversarial approaches](https://mlsanity.com/paper/2504.04751/) (2025)
- [Automatic DJ Transitions with Differentiable Audio Effects and Generative Adversarial Networks](https://mlsanity.com/paper/2110.06525/) (2021)
- [Comparative Study of State-based Neural Networks for Virtual Analog Audio Effects Modeling](https://mlsanity.com/paper/2405.04124/) (2024)
- [Automatic Equalization for Individual Instrument Tracks Using Convolutional Neural Networks](https://mlsanity.com/paper/2407.16691/) (2024)
- [Style Transfer for Non-differentiable Audio Effects](https://mlsanity.com/paper/2309.17125/) (2023)

## Abstract (arXiv)

In this work, we propose using differentiable cascaded biquads to model an audio distortion effect. We extend trainable infinite impulse response (IIR) filters to the hyperconditioned case, in which a transformation is learned to directly map external parameters of the distortion effect to its internal filter and gain parameters, along with activations necessary to ensure filter stability. We propose a novel, efficient training scheme of IIR filters by means of a Fourier transform. Our models have significantly fewer parameters and reduced complexity relative to more traditional black-box neural audio effect modeling methodologies using finite impulse response filters. Our smallest, best-performing model adequately models a BOSS MT-2 pedal at 44.1 kHz, using a total of 40 biquads and only 210 parameters. Its model parameters are interpretable, can be related back to the original analog audio circuit, and can even be intuitively altered by machine learning non-specialists after model training. Quantitative and qualitative results illustrate the effectiveness of the proposed method.