Evaluation of emotional arousal level and depression severity using the centripetal force derived from voice

In this research, we propose a new voice feature called centripetal force (CF) to investigate the relationship between emotional arousal level and depression severity. First, CFs were calculated from various speech recordings in the interactive emotional dyadic motion capture database, and the correlation with the arousal level of each voice was examined. The resulting correlation coefficient was 0.52. We collected a total of 178 datasets comprising 10 speech phrases and the Hamilton Rating Scale for Depression (HAM-D) score of outpatients with major depression at the Ginza Taimei Clinic (GTC) and the National Defense Medical College (NDMC) Hospital. The correlation coefficients between CF and HAM-D scores were -0.33 and -0.43 at the GTC and NDMC, respectively. Next, the dataset was divided into the no depression group (HAM-D<8) and the depression group (HAM-D [≥] 8) according to the HAM-D score. There was a significant difference in the mean CF values between the two groups in both the GTC and NDMC data (p = 0.0089 and p = 0.0016, respectively). The AUC when discriminating both groups by CF was 0.76 in GTC data and 0.72 in NDMC data. Indirectly, using CF established a relationship between emotional arousal level and depression severity.