# An Interpretable, Controllable Time-Varying IIR Denoiser for On-Device Assistive Hearing

arXiv: https://arxiv.org/abs/2603.02794 · HTML: https://mlsanity.com/paper/2603.02794/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 43 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Builds on (most cited first, 29 of 29)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [SDR - half-baked or well done?](https://mlsanity.com/paper/1811.02508/) (2018)
- [DNSMOS: A Non-Intrusive Perceptual Objective Speech Quality metric to evaluate Noise Suppressors](https://mlsanity.com/paper/2010.15258/) (2020)
- [Speech Enhancement and Dereverberation with Diffusion-based Generative Models](https://mlsanity.com/paper/2208.05830/) (2022)
- [DDSP: Differentiable Digital Signal Processing](https://mlsanity.com/paper/2001.04643/) (2020)
- [DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement](https://mlsanity.com/paper/2008.00264/) (2020)
- [The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Testing Framework, and Challenge Results](https://mlsanity.com/paper/2005.13981/) (2020)
- [DeepFilterNet: A Low Complexity Speech Enhancement Framework for Full-Band Audio based on Deep Filtering](https://mlsanity.com/paper/2110.05588/) (2021)
- [A Hybrid DSP/Deep Learning Approach to Real-Time Full-Band Speech Enhancement](https://mlsanity.com/paper/1709.08243/) (2017)
- [SAM Audio: Segment Anything in Audio](https://mlsanity.com/paper/2512.18099/) (2025)
- [Towards efficient models for real-time deep noise suppression](https://mlsanity.com/paper/2101.09249/) (2021)
- [FullSubNet+: Channel Attention FullSubNet with Complex Spectrograms for Speech Enhancement](https://mlsanity.com/paper/2203.12188/) (2022)
- [DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement](https://mlsanity.com/paper/2305.08227/) (2023)
- [DeepFilterNet2: Towards Real-Time Speech Enhancement on Embedded Devices for Full-Band Audio](https://mlsanity.com/paper/2205.05474/) (2022)
- [DCCRN+: Channel-wise Subband DCCRN with SNR Estimation for Speech Enhancement](https://mlsanity.com/paper/2106.08672/) (2021)
- [Style Transfer of Audio Effects with Differentiable Signal Processing](https://mlsanity.com/paper/2207.08759/) (2022)
- [ICASSP 2024 Speech Signal Improvement Challenge](https://mlsanity.com/paper/2401.14444/) (2024)
- [S-DCCRN: Super Wide Band DCCRN with learnable complex feature for speech enhancement](https://mlsanity.com/paper/2111.08387/) (2021)
- [CLCNet: Deep learning-based Noise Reduction for Hearing Aids using Complex Linear Coding](https://mlsanity.com/paper/2001.10218/) (2020)
- [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)
- [Low bit rate binaural link for improved ultra low-latency low-complexity multichannel speech enhancement in Hearing Aids](https://mlsanity.com/paper/2307.08858/) (2023)
- [High-Fidelity Noise Reduction with Differentiable Signal Processing](https://mlsanity.com/paper/2310.11364/) (2023)
- [DMF-Net: A decoupling-style multi-band fusion model for full-band speech enhancement](https://mlsanity.com/paper/2203.00472/) (2022)
- [Meta-AF: Meta-Learning for Adaptive Filters](https://mlsanity.com/paper/2204.11942/) (2022)
- [Ultra-lightweight Neural Differential DSP Vocoder For High Quality Speech Synthesis](https://mlsanity.com/paper/2401.10460/) (2024)
- [Improving Resource-Efficient Speech Enhancement via Neural Differentiable DSP Vocoder Refinement](https://mlsanity.com/paper/2508.14709/) (2025)
- [Accelerating Automatic Differentiation of Direct Form Digital Filters](https://mlsanity.com/paper/2511.14390/) (2025)

## Abstract (arXiv)

We present TVBC (Time-Varying Biquad Cascade), an interpretable, low-latency speech enhancement model for real-time, on-device assistive hearing. A lightweight neural controller predicts, in real time, the coefficients of a differentiable cascade of 35 second-order IIR filters (biquads), so the model tracks non-stationary noise while keeping a fully interpretable processing chain: every spectral modification is an explicit, adjustable equalizer curve rather than an opaque `black-box' transform. Because the biquad cascade carries the signal processing, the controller can be made very small, driving the cascade with only 24k parameters at a 10.7ms algorithmic latency, within hearing-aid budgets, and running entirely on-device so that audio never leaves the device. We also expose the suppression-versus-preservation trade-off as an explicit control: it can be set during training through the loss weighting, and adjusted at inference, with no retraining, by mixing the noisy input with the denoised output. On hearing-aid metrics (HASPI/HASQI) the 24k model stays within about 0.02 of DFNet3 (2.3M parameters, almost two orders of magnitude larger) while using about 29X fewer multiply-accumulates, although larger black-box models still lead on reference metrics such as PESQ. We present TVBC as a proof of concept for a compact, interpretable, and controllable denoiser for on-device assistive hearing.