PSLA

Reported on 3 benchmarks across 3 tasks · 1 paper · 3 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio2 results

Audio ClassificationonFSD50K
mAP· uses extra data· 2021-02-02
56.71
best: 69.7 (ONE-PEACE)
SOTA
PSLA: Improving Audio Tagging with Pretraining, Sampling, Labeling, and Aggregation arXiv:2102.01243
Audio TaggingonAudioSet
mean average precision· uses extra data· 2021-02-02
0.474
best: 0.512 (CAV-MAE (Audio-Visual))
SOTA
PSLA: Improving Audio Tagging with Pretraining, Sampling, Labeling, and Aggregation arXiv:2102.01243

Methodology1 result

ClassificationonFSD50K
mAP· uses extra data· 2021-02-02
56.71
best: 69.7 (ONE-PEACE)
SOTA
PSLA: Improving Audio Tagging with Pretraining, Sampling, Labeling, and Aggregation arXiv:2102.01243