Fig. 3From: Depression-level assessment from multi-lingual conversational speech data using acoustic and text featuresNumber of samples for each class after the clustering algorithm is applied. Turkish database is usedBack to article page