# Acoustic Scene Classification on UAS DCASE Task 1A 2018 (test)

Source: [Acoustic Scene Classification on UAS DCASE Task 1A 2018 (test)](<https://www.sota2.com/research/sota/acoustic-scene-classification-on-uas-dcase-task-1a-2018-test>)

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

Data updated: 2026-02-26T05:28:36.758Z

Evaluation of Acoustic Scene Classification performance on the test split of the DCASE 2018 Task 1A challenge.

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 | Acoustic Scene Classification |
| Dataset (including any stated split) | UAS DCASE Task 1A 2018 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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/acoustic-scene-classification-on-uas-dcase-task-1a-2018-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ERGL | 78.1 | Not recorded | 68fa6dda1e209131cea3c70e |
| ERGL | 78.08 | Not recorded | 68fa6dda1e209131cea3c70e |
| Relation-guided ASC (RGASC) | 77.35 | Not recorded | 68fa6dda1e209131cea3c70e |
| Wavelet-based spectrum | 76.6 | Not recorded | 68fa6dda1e209131cea3c70e |
| MLTF | 75.3 | Not recorded | 68fa6dda1e209131cea3c70e |
| PANN | 73.8 | Not recorded | 68fa6dda1e209131cea3c70e |
| ABCNN | 72.6 | Not recorded | 68fa6dda1e209131cea3c70e |
| Model with Attention | 70.8 | Not recorded | 68fa6dda1e209131cea3c70e |
| NNF\_CNNEns | 69.3 | Not recorded | 68fa6dda1e209131cea3c70e |
| CNN\_from\_Surrey | 68 | Not recorded | 68fa6dda1e209131cea3c70e |
| Conditional scene and event recognition | 66.39 | Not recorded | 68fa6dda1e209131cea3c70e |
| MTL-based event and scene analysis | 61.69 | Not recorded | 68fa6dda1e209131cea3c70e |
| Baseline | 59.7 | Not recorded | 68fa6dda1e209131cea3c70e |
| PANN | 56.9 | Not recorded | 68fa6dda1e209131cea3c70e |
| Scene and event jointly classification | 52.35 | Not recorded | 68fa6dda1e209131cea3c70e |

### ERGL

Accuracy: 78.1. Result ID: 698fec1f3a6c314849fccf56.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Implementation](<https://github.com/Yuanbo2020/ERGL>)

Recorded evaluation configuration: {"Model structure":"CNN and Graph Learning"}. This record does not assert independent reproduction.

### ERGL

Accuracy: 78.08. Result ID: 698fec1f3a6c314849fccf4c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Implementation](<https://github.com/Yuanbo2020/ERGL>)

Recorded evaluation configuration: {"Approach":"one-way event-to-scene inference"}. This record does not assert independent reproduction.

### Relation-guided ASC (RGASC)

Accuracy: 77.35. Result ID: 698fec1f3a6c314849fccf4b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c7b1e209131cea34362.md>)

Recorded evaluation configuration: {"Model structure":"two-tower model"}. This record does not assert independent reproduction.

### Wavelet-based spectrum

Accuracy: 76.6. Result ID: 698fec1f3a6c314849fccf55.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

Recorded evaluation configuration: {"Model structure":"CRNN"}. This record does not assert independent reproduction.

### MLTF

Accuracy: 75.3. Result ID: 698fec1f3a6c314849fccf54.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

Recorded evaluation configuration: {"Model structure":"CNN and SVM"}. This record does not assert independent reproduction.

### PANN

Accuracy: 73.8. Result ID: 698fec1f3a6c314849fccf53.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68351e209131cea1ac9d.md>)

[Implementation](<https://github.com/qiuqiangkong/audioset_tagging_cnn>)

Recorded evaluation configuration: {"Model structure":"VGG-like CNN","Transfer learning mode":"Fine-tuning mode"}. This record does not assert independent reproduction.

### ABCNN

Accuracy: 72.6. Result ID: 698fec1f3a6c314849fccf52.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

Recorded evaluation configuration: {"Model structure":"Attention-Based CNN"}. This record does not assert independent reproduction.

### Model with Attention

Accuracy: 70.8. Result ID: 698fec1f3a6c314849fccf51.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

Recorded evaluation configuration: {"Model structure":"CRNN with Self-attention"}. This record does not assert independent reproduction.

### NNF\_CNNEns

Accuracy: 69.3. Result ID: 698fec1f3a6c314849fccf50.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

Recorded evaluation configuration: {"Model structure":"CNN and nearest neighbor filters"}. This record does not assert independent reproduction.

### CNN\_from\_Surrey

Accuracy: 68. Result ID: 698fec1f3a6c314849fccf4f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ce1e209131cea127d2.md>)

Recorded evaluation configuration: {"Model structure":"CNN"}. This record does not assert independent reproduction.

### Conditional scene and event recognition

Accuracy: 66.39. Result ID: 698fec1f3a6c314849fccf4a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

Recorded evaluation configuration: {"Loss type":"conditional loss"}. This record does not assert independent reproduction.

### MTL-based event and scene analysis

Accuracy: 61.69. Result ID: 698fec1f3a6c314849fccf49.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa695c1e209131cea21cbf.md>)

Recorded evaluation configuration: {"Framework":"Multi-task learning"}. This record does not assert independent reproduction.

### Baseline

Accuracy: 59.7. Result ID: 698fec1f3a6c314849fccf4e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64e31e209131cea0719d.md>)

Recorded evaluation configuration: {"Model structure":"CNN"}. This record does not assert independent reproduction.

### PANN

Accuracy: 56.9. Result ID: 698fec1f3a6c314849fccf4d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68351e209131cea1ac9d.md>)

[Implementation](<https://github.com/qiuqiangkong/audioset_tagging_cnn>)

Recorded evaluation configuration: {"Model structure":"VGG-like CNN","Transfer learning mode":"Fixed mode"}. This record does not assert independent reproduction.

### Scene and event jointly classification

Accuracy: 52.35. Result ID: 698fec1f3a6c314849fccf48.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6dda1e209131cea3c70e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65f41e209131cea0d63b.md>)

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

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