# Training DeepFilterNet with Accurate Room Acoustic Simulations Improves Single-Channel Speech Enhancement

arXiv: https://arxiv.org/abs/2608.20971 · HTML: https://mlsanity.com/paper/2608.20971/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 40 references that are in the map.
- First posted on arXiv 21 Aug 2026.

## Builds on (most cited first, 13 of 13)

- [NeMo: a toolkit for building AI applications using Neural Modules](https://mlsanity.com/paper/1909.09577/) (2019)
- [Supervised Speech Separation Based on Deep Learning: An Overview](https://mlsanity.com/paper/1708.07524/) (2017)
- [WHAMR!: Noisy and Reverberant Single-Channel Speech Separation](https://mlsanity.com/paper/1910.10279/) (2019)
- [ICASSP 2022 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2202.13288/) (2022)
- [DeepFilterNet: A Low Complexity Speech Enhancement Framework for Full-Band Audio based on Deep Filtering](https://mlsanity.com/paper/2110.05588/) (2021)
- [GWA: A Large High-Quality Acoustic Dataset for Audio Processing](https://mlsanity.com/paper/2204.01787/) (2022)
- [DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement](https://mlsanity.com/paper/2305.08227/) (2023)
- [Dual-Signal Transformation LSTM Network for Real-Time Noise Suppression](https://mlsanity.com/paper/2005.07551/) (2020)
- [A study on more realistic room simulation for far-field keyword spotting](https://mlsanity.com/paper/2006.02774/) (2020)
- [How to (virtually) train your speaker localizer](https://mlsanity.com/paper/2211.16958/) (2022)
- [MB-RIRs: a Synthetic Room Impulse Response Dataset with Frequency-Dependent Absorption Coefficients](https://mlsanity.com/paper/2507.09750/) (2025)
- [Synthetic Wave-Geometric Impulse Responses for Improved Speech Dereverberation](https://mlsanity.com/paper/2212.05360/) (2022)
- [Egonoise Resilient Source Localization and Speech Enhancement for Drones Using a Hybrid Model and Learning-Based Approach](https://mlsanity.com/paper/2508.06310/) (2025)

## Built on by (most cited first, 1 of 1)

- [FFASR: Benchmarking Far-Field Automatic Speech Recognition using High-Fidelity Simulated RIRs](https://mlsanity.com/paper/2609.38897/) (2026)

## Abstract (arXiv)

We investigate how the realism of synthetic room impulse response (RIR) datasets affects the training of DeepFilterNet3 for single-channel speech enhancement. We compare a DNS4 image-source-method (ISM) RIR dataset with a higher-acoustic-fidelity dataset generated using hybrid wave-based and geometrical acoustics simulation. Rather than isolating individual simulation factors, we compare complete RIR generation pipelines while keeping the enhancement model unchanged. Models are evaluated on unseen measured RIRs using objective speech enhancement metrics and downstream automatic speech recognition (ASR). Training with the higher-fidelity dataset consistently yields modest improvements in objective metrics and substantially lower ASR word error rates than the ISM dataset. Although the experiments do not attribute these gains to individual modelling components, they show that increasing the overall realism of synthetic acoustic training data improves the generalization of DeepFilterNet3 to unseen measured environments.