What is Audio denoising and dereverberation?

Audio denoising and dereverberation Choose model and processing mode Cost, time, formats and parameters depend on the selected model.

Who is Audio denoising and dereverberation for?

Separation & stem split Audio denoising and dereverberation Choose model and processing mode Cost, time, formats and parameters depend on the selected model.

What tasks can Audio denoising and dereverberation help with?

Separation & stem split; A cleaned full mix reconstructed from XLANCE-MSR restored sources.; Cleaning a complete mix before mastering, separation, transcription, or video remux.; Music audio or video with noise, room reverb, compression artifacts, or livestream degradation.; Remove noise and room reverberation from music; Prepare a cleaner mix for separation or MIDI transcription

Audio denoising and dereverberation

Q: Which models and related terms are associated with Audio denoising and dereverberation?

XLANCE-MSR, XLANCE-MSR GitHub, XLANCE-MSR MIT checkpoints, MSR Challenge 2025 results, XLANCE-MSR arXiv, MSR Challenge summary, MSR Challenge 2025 official results, xlancelab

Models, parameters, and sources

XLANCE-MSRXLANCE-MSR GitHubXLANCE-MSR MIT checkpointsMSR Challenge 2025 resultsXLANCE-MSR arXivMSR Challenge summaryMSR Challenge 2025 official resultsxlancelabOfficial challenge resultOfficial challenge resultsCUPAudioGroupHachimi

XLANCE-MSR

XLANCE-MSR extends a six-stem BS-RoFormer system into eight-class source restoration with separation, denoising, and dereverberation/restoration. The MIT checkpoint weights are public.

XLANCE-MSR GitHub XLANCE-MSR MIT checkpoints MSR Challenge 2025 results XLANCE-MSR arXiv MSR Challenge summary

Targets eight source classes: vocals, guitar, keys, synth, bass, drums, percussion, and orchestral.
Official MSR Challenge 2025 results report objective rank #1 with MMSNR 4.4623, Zimt 0.0137, FAD 0.1988, and subjective overall MOS 3.4665. These are source-restoration metrics, not six-stem SDR.
Writes WAV stems only; the sound can be drier and closer to untreated material rather than preserving all record reverb, compression, EQ, mastering, and codec artifacts.

Model comparison

Model

Role

Why it is used

Output impact

Scores apply only to the cited dataset, metric and model version. Leading this table does not mean a current world record or a result from this service. Capabilities and tests can justify prioritizing models without comparable scores.

Benchmark chart: MSR Challenge 2025 source-restoration result / Music Denoise · MelBand RoFormer / Vocal De-Echo & Dereverb / Crowd Removal

MSR Challenge 2025 source-restoration result

MSR Challenge 2025 official results

Official results rank XLANCE-MSR first on objective metrics and listening tests; these are eight-class source-restoration scores, not six-stem SDR scores.

Public sources checked: 2026-09-10

Test setting	Model	MMSNR	Detailed metrics
Official challenge result	xlancelabOfficial challenge results	4.4623	Zimt0.0137FAD0.1988Separation MOS4.2358Restoration MOS3.3892Overall MOS3.4665
Official challenge result	CUPAudioGroupOfficial challenge results	2.3405	Zimt0.0164FAD0.2253Separation MOS3.8360Restoration MOS2.9173Overall MOS2.9253
Official challenge result	HachimiOfficial challenge results	2.0016	Zimt0.0183FAD0.2939Separation MOS3.5814Restoration MOS2.6331Overall MOS2.7235
Official challenge result	AC_DCOfficial challenge results	1.4520	Zimt0.0182FAD0.2907Separation MOS3.5425Restoration MOS2.4768Overall MOS2.5412
Official challenge result	cp-jkuOfficial challenge results	0.8329	Zimt0.0189FAD0.3814Separation MOS3.5510Restoration MOS2.0838Overall MOS2.1414

Music Denoise · MelBand RoFormer

Checkpoint-metadata SDR is parsed from the official released filename. It is not a new TelkNet measurement and must not be compared across different datasets or protocols.

Public sources checked: 2026-09-10

Denoise MelBand RoFormer

denoise_mel_band_roformer_aufr33_sdr_27.9959.ckpt

Denoise MelBand RoFormer · Aggressive

denoise_mel_band_roformer_aufr33_aggr_sdr_27.9768.ckpt

Vocal De-Echo & Dereverb

Checkpoint-metadata SDR is parsed from the official released filename. It is not a new TelkNet measurement and must not be compared across different datasets or protocols.

Public sources checked: 2026-09-10

Vocal De-Echo MelBand RoFormer

dereverb_echo_mbr_v2_sdr_dry_13.4843.ckpt

Crowd Removal

Checkpoint-metadata SDR is parsed from the official released filename. It is not a new TelkNet measurement and must not be compared across different datasets or protocols.

Public sources checked: 2026-09-10

Crowd MelBand RoFormer

mel_band_roformer_crowd_aufr33_viperx_sdr_8.7144.ckpt

Paper and benchmark notes

XLANCE-MSR Technical Report for MSR Challenge 2025

arXiv - 2026

Technical report for the XLANCE-MSR system submitted to the first Music Source Restoration Challenge.

arXiv:2602.09042 Official challenge page

A Summary of the First Music Source Restoration Challenge

arXiv - 2026

Challenge summary paper describing the MSR task, datasets, objective metrics, subjective evaluation, and final ranking tables.

arXiv:2601.04343 MSR Challenge

Official / repositories

XLANCE-MSR GitHub XLANCE-MSR MIT checkpoints MSR Challenge 2025 results MSST inference source (pinned)Official denoise checkpoint snapshot Official aggressive denoise checkpoint snapshot Official de-echo checkpoint (pinned)Official crowd checkpoint snapshot

Papers / technical notes

XLANCE-MSR arXiv MSR Challenge summary

Parameter guide

Official MSST test-shift count. Higher values cost more time and may improve stability.
Official test-time augmentation. It increases processing time and is disabled by default.

Audio denoising and dereverberation

Models, parameters, and sources

XLANCE-MSR

Denoise MelBand RoFormer

Denoise MelBand RoFormer · Aggressive

Vocal De-Echo MelBand RoFormer

Crowd MelBand RoFormer

Model comparison

Benchmark chart: MSR Challenge 2025 source-restoration result / Music Denoise · MelBand RoFormer / Vocal De-Echo & Dereverb / Crowd Removal

MSR Challenge 2025 source-restoration result

Music Denoise · MelBand RoFormer

Vocal De-Echo & Dereverb

Crowd Removal

Paper and benchmark notes

XLANCE-MSR Technical Report for MSR Challenge 2025

A Summary of the First Music Source Restoration Challenge

Official / repositories

Papers / technical notes

Parameter guide

Workflow

Before you submit

Quick feedback