# Audio Source Separation on AudioSet SingleSource (test)

Source: [Audio Source Separation on AudioSet SingleSource (test)](<https://www.sota2.com/research/sota/audio-source-separation-on-audioset-singlesource-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:02:50.126Z

Evaluation of Audio Source Separation performance on the test split of the AudioSet SingleSource dataset, typically reporting metrics like SDR, SIR, and SAR.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Audio Source Separation |
| Dataset (including any stated split) | AudioSet SingleSource (test) |
| Metric | SDR |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [SDR](<https://www.sota2.com/research/sota/audio-source-separation-on-audioset-singlesource-test.md?metric=SDR>): Higher is better; unit: SCALAR.

- [SIR](<https://www.sota2.com/research/sota/audio-source-separation-on-audioset-singlesource-test.md?metric=SIR>): Higher is better; unit: SCALAR.

- [SAR](<https://www.sota2.com/research/sota/audio-source-separation-on-audioset-singlesource-test.md?metric=SAR>): Higher is better; unit: SCALAR.

## Recorded results

| Method | SDR | Result date | Report paper |
| --- | --- | --- | --- |
| CO-SEPARATION | 4.26 | Not recorded | 68fa66381e209131cea0edfa |
| AV-MIML | 1.83 | Not recorded | 68fa66381e209131cea0edfa |
| AV-Mix-and-Separate | 1.68 | Not recorded | 68fa66381e209131cea0edfa |
| Sound-of-Pixels | 1.66 | Not recorded | 68fa66381e209131cea0edfa |
| NMF-MFCC | 0.25 | Not recorded | 68fa66381e209131cea0edfa |

### CO-SEPARATION

SDR: 4.26. Result ID: 698ce1df8b517eb4331d66e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0edfa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0edfa.md>)

[Implementation](<https://github.com/hangzhaomit/Sound-of-Pixels>)

Recorded evaluation configuration: {"training":"AudioSet-Unlabeled"}. This record does not assert independent reproduction.

### AV-MIML

SDR: 1.83. Result ID: 698ce1df8b517eb4331d66e0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0edfa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa649b1e209131cea0584a.md>)

Recorded evaluation configuration: {"training":"AudioSet-Unlabeled"}. This record does not assert independent reproduction.

### AV-Mix-and-Separate

SDR: 1.68. Result ID: 698ce1df8b517eb4331d66e1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0edfa.md>)

Recorded evaluation configuration: {"training":"AudioSet-Unlabeled"}. This record does not assert independent reproduction.

### Sound-of-Pixels

SDR: 1.66. Result ID: 698ce1df8b517eb4331d66df.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0edfa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64f91e209131cea077c6.md>)

Recorded evaluation configuration: {"training":"AudioSet-Unlabeled"}. This record does not assert independent reproduction.

### NMF-MFCC

SDR: 0.25. Result ID: 698ce1df8b517eb4331d66e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0edfa.md>)

Recorded evaluation configuration: {"type":"non-learned"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
