mlsanityData to 14 Oct 2026

MB-RIRs: a Synthetic Room Impulse Response Dataset with Frequency-Dependent Absorption Coefficients

arXiv 2507.09750 ↗First posted 13 Jul 2025cs.SD cs.LG eess.AScited by 3

Not on the map. Why?

Abstract (arXiv)

We investigate the effects of four strategies for improving the ecological validity of synthetic room impulse response (RIR) datasets for monoaural Speech Enhancement (SE). We implement three features on top of the traditional image source method-based (ISM) shoebox RIRs: multiband absorption coefficients, source directivity and receiver directivity. We additionally consider mesh-based RIRs from the SoundSpaces dataset. We then train a DeepFilternet3 model for each RIR dataset and evaluate the performance on a test set of real RIRs both objectively and subjectively. We find that RIRs which use frequency-dependent acoustic absorption coefficients (MB-RIRs) can obtain +0.51dB of SDR and a +8.9 MUSHRA score when evaluated on real RIRs. The MB-RIRs dataset is publicly available for free download.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 18 of its 33 references that are in the map. First posted on arXiv 13 Jul 2025.

Builds on

18 of 33 references.

References · 18

  1. 2020DNSMOS: A Non-Intrusive Perceptual Objective Speech Quality metric to evaluate Noise Suppressorscited by 241
  2. 2022Speech Enhancement and Dereverberation with Diffusion-based Generative Modelscited by 197
  3. 2021NISQA: A Deep CNN-Self-Attention Model for Multidimensional Speech Quality Prediction with Crowdsourced Datasetscited by 192
  4. 2022SoundSpaces 2.0: A Simulation Platform for Visual-Acoustic Learningcited by 95
  5. 2023ICASSP 2023 Deep Noise Suppression Challengecited by 65
  6. 2023TorchAudio-Squim: Reference-less Speech Quality and Intelligibility measures in TorchAudiocited by 59
  7. 2018Building and Evaluation of a Real Room Impulse Response Datasetcited by 43
  8. 2022GWA: A Large High-Quality Acoustic Dataset for Audio Processingcited by 25
  9. 2023DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancementcited by 25
  10. 2024URGENT Challenge: Universality, Robustness, and Generalizability For Speech Enhancementcited by 17
  11. 2020An empirical study of Conv-TasNetcited by 15
  12. 2021TS-RIR: Translated synthetic room impulse responses for speech augmentationcited by 12
6 more
  1. 2023Toward Universal Speech Enhancement for Diverse Input Conditionscited by 10
  2. 2025Generative Data Augmentation Challenge: Synthesis of Room Acoustics for Speaker Distance Estimationcited by 9
  3. 2024Why does music source separation benefit from cacophony?cited by 8
  4. 2020A study on more realistic room simulation for far-field keyword spottingcited by 4
  5. 2024Overview of the L3DAS23 Challenge on Audio-Visual Extended Realitycited by 4
  6. 2023An objective evaluation of Hearing Aids and DNN-based speech enhancement in complex acoustic scenescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Treble10: A high-quality dataset for far-field speech recognition, dereverberation, and enhancementcited by 2
  2. 2026Improving multichannel speech enhancement through accurate room-acoustic simulationscited by 1
  3. 2026Training DeepFilterNet with Accurate Room Acoustic Simulations Improves Single-Channel Speech Enhancementcited by 1