# Audio Tagging on MTT

Source: [Audio Tagging on MTT](<https://www.sota2.com/research/sota/audio-tagging-on-mtt>)

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

Data updated: 2026-04-29T07:02:54.506Z

Evaluation of audio tagging performance using the MTT dataset, measured by MTT-AP and AUC.

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 Tagging |
| Dataset (including any stated split) | MTT |
| Metric | MTT AP |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 19 of 19 |
| 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

- [MTT AP](<https://www.sota2.com/research/sota/audio-tagging-on-mtt.md?metric=MTT%20AP>): Higher is better; unit: PERCENT.

- [AUC](<https://www.sota2.com/research/sota/audio-tagging-on-mtt.md?metric=AUC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | MTT AP | Result date | Report paper |
| --- | --- | --- | --- |
| MATPAC++ | 40.8 | Not recorded | 69f1907155a3c32af89fb976 |
| M2D | 40.6 | Not recorded | 69f1907155a3c32af89fb976 |
| S-SONDO (MobileNetV3) | 40.2 | Not recorded | 69f1907155a3c32af89fb976 |
| S-SONDO (DyMN) | 39.9 | Not recorded | 69f1907155a3c32af89fb976 |
| S-SONDO (MobileNetV3) | 39.5 | Not recorded | 69f1907155a3c32af89fb976 |
| S-SONDO (ERes2Net) | 39.4 | Not recorded | 69f1907155a3c32af89fb976 |
| S-SONDO (DyMN) | 39.1 | Not recorded | 69f1907155a3c32af89fb976 |
| DyMN | 38.8 | Not recorded | 69f1907155a3c32af89fb976 |
| S-SONDO (ERes2Net) | 38.8 | Not recorded | 69f1907155a3c32af89fb976 |
| MobileNetV3 | 38.7 | Not recorded | 69f1907155a3c32af89fb976 |
| DuoTok | 35 | Not recorded | 692689a2d6b0dd33429304ad |
| ERes2Net | 34.3 | Not recorded | 69f1907155a3c32af89fb976 |
| SemantiCodec | 32 | Not recorded | 692689a2d6b0dd33429304ad |
| X-Codec | 32 | Not recorded | 692689a2d6b0dd33429304ad |
| X-Codec (YuE) | 32 | Not recorded | 692689a2d6b0dd33429304ad |
| MuCodec (LeVo) | 26 | Not recorded | 692689a2d6b0dd33429304ad |
| DAC | 20 | Not recorded | 692689a2d6b0dd33429304ad |
| EnCodec | 18 | Not recorded | 692689a2d6b0dd33429304ad |
| WavTokenizer | 17 | Not recorded | 692689a2d6b0dd33429304ad |

### MATPAC++

MTT AP: 40.8. Result ID: 69f1ad0d47f34ce924d7ab60.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7aaa1e209131cea847bb.md>)

Recorded evaluation configuration: {"Role":"Teacher","Size":"86M"}. This record does not assert independent reproduction.

### M2D

MTT AP: 40.6. Result ID: 69f1ad0d47f34ce924d7ab61.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea91e209131cea4130f.md>)

[Implementation](<https://github.com/nttcslab/m2d>)

Recorded evaluation configuration: {"Role":"Teacher","Size":"86M"}. This record does not assert independent reproduction.

### S-SONDO (MobileNetV3)

MTT AP: 40.2. Result ID: 69f1ad0d47f34ce924d7ab58.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

Recorded evaluation configuration: {"Teacher":"MATPAC++","Size":"2.9M"}. This record does not assert independent reproduction.

### S-SONDO (DyMN)

MTT AP: 39.9. Result ID: 69f1ad0d47f34ce924d7ab5b.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

Recorded evaluation configuration: {"Teacher":"MATPAC++","Size":"8.7M"}. This record does not assert independent reproduction.

### S-SONDO (MobileNetV3)

MTT AP: 39.5. Result ID: 69f1ad0d47f34ce924d7ab59.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

Recorded evaluation configuration: {"Teacher":"M2D","Size":"2.9M"}. This record does not assert independent reproduction.

### S-SONDO (ERes2Net)

MTT AP: 39.4. Result ID: 69f1ad0d47f34ce924d7ab5e.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

Recorded evaluation configuration: {"Teacher":"MATPAC++","Size":"1.4M"}. This record does not assert independent reproduction.

### S-SONDO (DyMN)

MTT AP: 39.1. Result ID: 69f1ad0d47f34ce924d7ab5c.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

Recorded evaluation configuration: {"Teacher":"M2D","Size":"8.7M"}. This record does not assert independent reproduction.

### DyMN

MTT AP: 38.8. Result ID: 69f1ad0d47f34ce924d7ab5a.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f11e209131cea4e300.md>)

[Implementation](<https://github.com/fschmid56/EfficientAT>)

Recorded evaluation configuration: {"Teacher":"None (Supervised)","Size":"8.7M"}. This record does not assert independent reproduction.

### S-SONDO (ERes2Net)

MTT AP: 38.8. Result ID: 69f1ad0d47f34ce924d7ab5f.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

Recorded evaluation configuration: {"Teacher":"M2D","Size":"1.4M"}. This record does not assert independent reproduction.

### MobileNetV3

MTT AP: 38.7. Result ID: 69f1ad0d47f34ce924d7ab57.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ac1e209131cea11957.md>)

Recorded evaluation configuration: {"Teacher":"None (Supervised)","Size":"2.9M"}. This record does not assert independent reproduction.

### DuoTok

MTT AP: 35. Result ID: 69ce371347f34ce924b7b225.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

Recorded evaluation configuration: {"Rate":"25 Hz","Size (S)":"2×32768","Kbps":"0.75"}. This record does not assert independent reproduction.

### ERes2Net

MTT AP: 34.3. Result ID: 69f1ad0d47f34ce924d7ab5d.

[Reporting paper](<https://www.sota2.com/research/paper/69f1907155a3c32af89fb976.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703f1e209131cea49ecd.md>)

Recorded evaluation configuration: {"Teacher":"None (Supervised)","Size":"1.4M"}. This record does not assert independent reproduction.

### SemantiCodec

MTT AP: 32. Result ID: 69ce371347f34ce924b7b221.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75ec1e209131cea6aa24.md>)

Recorded evaluation configuration: {"Rate":"50 Hz","Size (S)":"2×8192","Kbps":"1.30"}. This record does not assert independent reproduction.

### X-Codec

MTT AP: 32. Result ID: 69ce371347f34ce924b7b222.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e61e209131cea6a88b.md>)

Recorded evaluation configuration: {"Rate":"50 Hz","Size (S)":"8×1024","Kbps":"4.00"}. This record does not assert independent reproduction.

### X-Codec (YuE)

MTT AP: 32. Result ID: 69ce371347f34ce924b7b223.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b451e209131cea876fc.md>)

Recorded evaluation configuration: {"Rate":"50 Hz","Size (S)":"16×1024","Kbps":"8.00"}. This record does not assert independent reproduction.

### MuCodec (LeVo)

MTT AP: 26. Result ID: 69ce371347f34ce924b7b224.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79731e209131cea7db13.md>)

Recorded evaluation configuration: {"Rate":"25 Hz","Size (S)":"2×16384","Kbps":"0.70"}. This record does not assert independent reproduction.

### DAC

MTT AP: 20. Result ID: 69ce371347f34ce924b7b21e.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71091e209131cea4ecde.md>)

[Implementation](<https://github.com/descriptinc/descript-audio-codec>)

Recorded evaluation configuration: {"Rate":"75 Hz","Size (S)":"8×1024","Kbps":"6.00"}. This record does not assert independent reproduction.

### EnCodec

MTT AP: 18. Result ID: 69ce371347f34ce924b7b21f.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6db91e209131cea3b969.md>)

Recorded evaluation configuration: {"Rate":"75 Hz","Size (S)":"8×1024","Kbps":"6.00"}. This record does not assert independent reproduction.

### WavTokenizer

MTT AP: 17. Result ID: 69ce371347f34ce924b7b220.

[Reporting paper](<https://www.sota2.com/research/paper/692689a2d6b0dd33429304ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa771b1e209131cea7109d.md>)

Recorded evaluation configuration: {"Rate":"40 Hz","Size (S)":"1×4096","Kbps":"0.48"}. This record does not assert independent reproduction.

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