Fig. 6From: Depression-level assessment from multi-lingual conversational speech data using acoustic and text featuresDistribution of squared errors for the baseline MRMR case is shown in the top figure for the Turkish task. The middle figure shows the squared error distribution for the baseline MRMR case after fusion. Bottom figure shows the squared error distribution of the ml-MRMR system with Nmin=3 with English and after fusionBack to article page