Table 4. F1-score of machine learning models using different colony sound features on three colony sound datasets.

Datasets	Dataset 1				Dataset 2				Dataset 3
Algorithm	KNN	DT	RF	SVM	KNN	DT	RF	SVM	KNN	DT	RF	SVM
VGGish	94.79%	93.45%	94.42%	91.55%	86.58%	85.17%	85.93%	81.49%	91.06%	88.82%	89.21%	89.03%
MFCC	68.24%	66.32%	68.49%	65.26%	90.13%	88.44%	89.63%	85.41%	65.73%	65.74%	64.80%	65.09%