Differentiable Time-Varying IIR Filtering for Real-Time Speech Denoising
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917309186048000 |
|---|---|
| author | Rota, Riccardo Ratmanski, Kiril Coldenhoff, Jozef Cernak, Milos |
| author_facet | Rota, Riccardo Ratmanski, Kiril Coldenhoff, Jozef Cernak, Milos |
| contents | We present TVF (Time-Varying Filtering), a low-latency speech enhancement model with 1 million parameters. Combining the interpretability of Digital Signal Processing (DSP) with the adaptability of deep learning, TVF bridges the gap between traditional filtering and modern neural speech modeling. The model utilizes a lightweight neural network backbone to predict the coefficients of a differentiable 35-band IIR filter cascade in real time, allowing it to adapt dynamically to non-stationary noise. Unlike ``black-box'' deep learning approaches, TVF offers a completely interpretable processing chain, where spectral modifications are explicit and adjustable. We demonstrate the efficacy of this approach on a speech denoising task using the Valentini-Botinhao dataset and compare the results to a static DDSP approach and a fully deep-learning-based solution, showing that TVF achieves effective adaptation to changing noise conditions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_02794 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Differentiable Time-Varying IIR Filtering for Real-Time Speech Denoising Rota, Riccardo Ratmanski, Kiril Coldenhoff, Jozef Cernak, Milos Sound Artificial Intelligence Machine Learning Audio and Speech Processing We present TVF (Time-Varying Filtering), a low-latency speech enhancement model with 1 million parameters. Combining the interpretability of Digital Signal Processing (DSP) with the adaptability of deep learning, TVF bridges the gap between traditional filtering and modern neural speech modeling. The model utilizes a lightweight neural network backbone to predict the coefficients of a differentiable 35-band IIR filter cascade in real time, allowing it to adapt dynamically to non-stationary noise. Unlike ``black-box'' deep learning approaches, TVF offers a completely interpretable processing chain, where spectral modifications are explicit and adjustable. We demonstrate the efficacy of this approach on a speech denoising task using the Valentini-Botinhao dataset and compare the results to a static DDSP approach and a fully deep-learning-based solution, showing that TVF achieves effective adaptation to changing noise conditions. |
| title | Differentiable Time-Varying IIR Filtering for Real-Time Speech Denoising |
| topic | Sound Artificial Intelligence Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2603.02794 |