Table 2. Results of the prediction models.

	Features	Dimensionality reduction	Classifier	Accuracy	Specificity	Sensitivity	AUC
Female group	Opensmile ComParE 2016 (6373)	200 selected features	LR	0.60 (0.03)	0.60 (0.03)	0.62 (0.07)	0.62 (0.02)
			MLP Classifier	0.63 (0.02)	0.61 (0.02)	0.74 (0.02)	0.66 (0.02)
			SVM RBF	0.57 (0.02)	0.57 (0.02)	0.63 (0.03)	0.61 (0.01)
	Byol-S embeddings (2048)	PCA, n_components = n_samples	LR	0.67 (0.04)	0.68 (0.04)	0.65 (0.11)	0.70 (0.06)
			MLP Classifier	0.67 (0.04)	0.66 (0.04)	0.67 (0.11)	0.71 (0.07)
			SVM RBF	0.66 (0.04)	0.65 (0.07)	0.67 (0.11)	0.71 (0.05)
Male group	Opensmile ComParE 2016 (6373)	100 selected features	LR	0.56 (0.02)	0.55 (0.01)	0.58 (0.05)	0.61 (0.05)
			MLP Classifier	0.61 (0.05)	0.61 (0.06)	0.63 (0.06)	0.64 (0.05)
			SVM RBF	0.57 (0.05)	0.57 (0.06)	0.54 (0.05)	0.57 (0.05)
	Byol-S embeddings (2048)	PCA, n_components = 100	LR	0.69 (0.04)	0.66 (0.07)	0.72 (0.03)	0.73 (0.06)
			MLP Classifier	0.71 (0.02)	0.70 (0.02)	0.73 (0.03)	0.75 (0.05)
			SVM RBF	0.70 (0.04)	0.64 (0.05)	0.76 (0.03)	0.78 (0.05)

Table 2 presents the mean and standard deviation (in parentheses) of the performance metrics across cross-validation folds. The selected algorithm for each gender group is highlighted in bold. Logistic Regression (LR), Multi-layer Perceptron (MLP), Support Vector Machine Radial basis function kernel (SVM RBF).