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

Name: MB-RIRs: a Synthetic Room Impulse Response Dataset with Frequency-Dependent Absorption Coefficients
Keywords: Machine Learning, FOS: Computer and information sciences, Sound (cs.SD), Sound, Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, Audio and Speech Processing, Machine Learning (cs.LG)

descriptionPublicationkeyboard_double_arrow_right Article , Preprint 01 Jan 2025Embargo end date: 01 Jan 2025Publisher:arXiv

Authors: Gusó, Enric; Luberadzka, Joanna; Sayin, Umut; Serra, Xavier;

doi: 10.48550/arxiv.2507.09750

arXiv: http://arxiv.org/abs/2507.09750

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

- Summary
- Subjects
- Metrics

Abstract

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.

Accepted to WASPAA25

Keywords

Machine Learning, FOS: Computer and information sciences, Sound (cs.SD), Sound, Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, Audio and Speech Processing, Machine Learning (cs.LG)

Impact byBIP!

	selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).	0
	popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.	Average
	influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).	Average
	impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.	Average

Found an issue? Give us feedback

Average

Green