Masked Latent Prediction and Classification for Self-Supervised Audio Representation Learning

Benchmark Model Rank Results
environmental-sound-classification-on-urbansound8kMATPAC (SSL, linear eval)#2Accuracy: 89.4
instrument-recognition-on-openmic-2018MATPAC (SSL Model, linear eval)#2mean average precision: 0.854
music-auto-tagging-on-magnatagatuneMATPAC (SSL, linear eval)#2PR-AUC: 41.1ROC AUC: 91.6
instrument-recognition-on-nsynthMATPAC (SSL, linear eval)#4Accuracy: 74.6
audio-classification-on-fsd50kMATPAC (SSL Model)#8mAP: 55.2
audio-classification-on-esc-50MATPAC (SSL model, linear eval)#17Top-1 Accuracy: 93.5PRE-TRAINING DATASET: AudioSet…