# Music Genre Classification on GTZAN (test)

Source: [Music Genre Classification on GTZAN (test)](<https://www.sota2.com/research/sota/music-genre-classification-on-gtzan-test>)

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

Data updated: 2026-04-23T10:51:23.539Z

Evaluation of Music Genre Classification performance on the test split of the GTZAN dataset, typically measured by accuracy.

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 | Music Genre Classification |
| Dataset (including any stated split) | GTZAN (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 27 of 27 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/music-genre-classification-on-gtzan-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| LAION-CLAP | 96 | Not recorded | 68fa737a1e209131cea5cc73 |
| Explicit Dropout (K) | 85.78 | Not recorded | 69e9a78755a3c32af89f700b |
| Explicit Dropout (Q) | 85.68 | Not recorded | 69e9a78755a3c32af89f700b |
| Arora et al. (Q) | 85.31 | Not recorded | 69e9a78755a3c32af89f700b |
| Arora et al. (V) | 85.16 | Not recorded | 69e9a78755a3c32af89f700b |
| Explicit Dropout (None) | 85.16 | Not recorded | 69e9a78755a3c32af89f700b |
| None (Implicit) | 85 | Not recorded | 69e9a78755a3c32af89f700b |
| Explicit Dropout (V) | 85 | Not recorded | 69e9a78755a3c32af89f700b |
| Explicit Dropout (AV) | 85 | Not recorded | 69e9a78755a3c32af89f700b |
| DropAttention | 84.84 | Not recorded | 69e9a78755a3c32af89f700b |
| DropKey | 84.84 | Not recorded | 69e9a78755a3c32af89f700b |
| Arora et al. (FF) | 84.84 | Not recorded | 69e9a78755a3c32af89f700b |
| Arora et al. (K) | 84.22 | Not recorded | 69e9a78755a3c32af89f700b |
| Proposed DNN classifier | 81.5 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| Proposed DNN classifier | 81.5 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| Proposed DNN classifier | 79.5 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| Proposed DNN classifier | 78 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| Proposed DNN classifier | 77 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| M2D-CLAP | 75.17 | Not recorded | 68fa737a1e209131cea5cc73 |
| Niizumi et al. | 70.1 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| Niizumi et al. | 70.1 | Not recorded | 69b8dff10636e5bf4afbe8b9 |
| WavCaps | 52.1 | Not recorded | 68fa737a1e209131cea5cc73 |
| JMLA | 50.3 | Not recorded | 68fa737a1e209131cea5cc73 |
| Proto-LC | 47.24 | Not recorded | 68fa737a1e209131cea5cc73 |
| AudioCLIP | 45.52 | Not recorded | 68fa737a1e209131cea5cc73 |
| Pengi | 28.97 | Not recorded | 68fa737a1e209131cea5cc73 |
| Wav2CLIP | 27.66 | Not recorded | 68fa737a1e209131cea5cc73 |

### LAION-CLAP

Accuracy: 96. Result ID: 6992e3853a6c314849122e91.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa729e1e209131cea57959.md>)

Recorded evaluation configuration: {"zero-shot":true}. This record does not assert independent reproduction.

### Explicit Dropout (K)

Accuracy: 85.78. Result ID: 69e9f99047f34ce924d1f730.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"Explicit (K)","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Explicit Dropout (Q)

Accuracy: 85.68. Result ID: 69e9f99047f34ce924d1f72f.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"Explicit (Q)","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Arora et al. (Q)

Accuracy: 85.31. Result ID: 69e9f99047f34ce924d1f72a.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67211e209131cea147b6.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"(Q) Arora et al. \[1\]","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Arora et al. (V)

Accuracy: 85.16. Result ID: 69e9f99047f34ce924d1f72c.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67211e209131cea147b6.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"(V) Arora et al. \[1\]","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Explicit Dropout (None)

Accuracy: 85.16. Result ID: 69e9f99047f34ce924d1f72e.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"None","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### None (Implicit)

Accuracy: 85. Result ID: 69e9f99047f34ce924d1f729.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"None","Dropout on FF network":"Implicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Explicit Dropout (V)

Accuracy: 85. Result ID: 69e9f99047f34ce924d1f731.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"Explicit (V)","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Explicit Dropout (AV)

Accuracy: 85. Result ID: 69e9f99047f34ce924d1f732.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"Explicit (AV)","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### DropAttention

Accuracy: 84.84. Result ID: 69e9f99047f34ce924d1f727.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66121e209131cea0e134.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"DropAttention","Dropout on FF network":"Implicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### DropKey

Accuracy: 84.84. Result ID: 69e9f99047f34ce924d1f728.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"DropKey","Dropout on FF network":"Implicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Arora et al. (FF)

Accuracy: 84.84. Result ID: 69e9f99047f34ce924d1f72d.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67211e209131cea147b6.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"None","Dropout on FF network":"Arora et al. \[1\]","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Arora et al. (K)

Accuracy: 84.22. Result ID: 69e9f99047f34ce924d1f72b.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78755a3c32af89f700b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67211e209131cea147b6.md>)

Recorded evaluation configuration: {"Dropout on Attention Sequence":"(K) Arora et al. \[1\]","Dropout on FF network":"Explicit","Dropout ratio":"0.2","Model Architecture":"2-layer Transformer encoder"}. This record does not assert independent reproduction.

### Proposed DNN classifier

Accuracy: 81.5. Result ID: 69b915e147f34ce924a1c1d2.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

Recorded evaluation configuration: {"Feature Extractor":"BYOL-A embeddings"}. This record does not assert independent reproduction.

### Proposed DNN classifier

Accuracy: 81.5. Result ID: 69b915e147f34ce924a1c1da.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

Recorded evaluation configuration: {"Training Dataset":"GTZAN"}. This record does not assert independent reproduction.

### Proposed DNN classifier

Accuracy: 79.5. Result ID: 69b915e147f34ce924a1c1d4.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

Recorded evaluation configuration: {"Feature Extractor":"VGGish embeddings"}. This record does not assert independent reproduction.

### Proposed DNN classifier

Accuracy: 78. Result ID: 69b915e147f34ce924a1c1db.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

Recorded evaluation configuration: {"Training Dataset":"GTZAN + FMA-Small"}. This record does not assert independent reproduction.

### Proposed DNN classifier

Accuracy: 77. Result ID: 69b915e147f34ce924a1c1d3.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

Recorded evaluation configuration: {"Feature Extractor":"PANNs embeddings"}. This record does not assert independent reproduction.

### M2D-CLAP

Accuracy: 75.17. Result ID: 6992e3853a6c314849122e95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

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

Recorded evaluation configuration: {"masking\_ratio":"0.7","zero-shot":true}. This record does not assert independent reproduction.

### Niizumi et al.

Accuracy: 70.1. Result ID: 69b915e147f34ce924a1c1d1.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e9d1e209131cea411ae.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Niizumi et al.

Accuracy: 70.1. Result ID: 69b915e147f34ce924a1c1d9.

[Reporting paper](<https://www.sota2.com/research/paper/69b8dff10636e5bf4afbe8b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e9d1e209131cea411ae.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### WavCaps

Accuracy: 52.1. Result ID: 6992e3853a6c314849122e90.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74311e209131cea6055f.md>)

Recorded evaluation configuration: {"zero-shot":true}. This record does not assert independent reproduction.

### JMLA

Accuracy: 50.3. Result ID: 6992e3853a6c314849122e94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70eb1e209131cea4e020.md>)

Recorded evaluation configuration: {"zero-shot":false,"note":"Used test task data during training"}. This record does not assert independent reproduction.

### Proto-LC

Accuracy: 47.24. Result ID: 6992e3853a6c314849122e92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70541e209131cea4a4bb.md>)

Recorded evaluation configuration: {"zero-shot":true}. This record does not assert independent reproduction.

### AudioCLIP

Accuracy: 45.52. Result ID: 6992e3853a6c314849122e8e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d6a1e209131cea39b2c.md>)

[Implementation](<https://github.com/AndreyGuzhov/AudioCLIP>)

Recorded evaluation configuration: {"zero-shot":false,"note":"Used test task data during training"}. This record does not assert independent reproduction.

### Pengi

Accuracy: 28.97. Result ID: 6992e3853a6c314849122e93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71d71e209131cea534a3.md>)

[Implementation](<https://github.com/microsoft/Pengi>)

Recorded evaluation configuration: {"zero-shot":true}. This record does not assert independent reproduction.

### Wav2CLIP

Accuracy: 27.66. Result ID: 6992e3853a6c314849122e8f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa737a1e209131cea5cc73.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bd81e209131cea3085f.md>)

Recorded evaluation configuration: {"zero-shot":true}. This record does not assert independent reproduction.

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