Human non-verbal vocal bursts are evolutionary \nconservative emotional expressions. Humans can \neasily assess inner states of conspecifics based on \nthese calls. Moreover, they can attribute emotions \nto non-human animal vocalizations too. However, \nwhether the same acoustic cues are used to assess \nemotional content in conspecific and nonconspecific \nvocalizations is not clarified yet. \nTo test this, we compiled a pool of 100-100 various \ndog and human non-verbal vocalizations from \ndiverse social contexts, and designed an online \nsurvey, in which every sample could be rated along \nemotional valence and intensity. We also measured \nwithin each sample the average length of calls, the \nfundamental frequency and the harmonics-to-noise \nratio. \nWhile valence ratings did not differ across species, \nhuman vocalizations were less intense. Linear \nregressions revealed that both shorter dog and \nhuman calls were rated as more positive. In \ncontrast, subjects scored higher pitched human and \ndog sounds to be more intense. We also found dog \nvocalizations with shorter call length or with higher \nHNR were rated less intense. \nIn conclusion, acoustical parameters affected \nhumans’ emotional ratings independently from the \nsource species of these vocalizations. These findings \nsuggest that humans utilize the same mental \nmechanisms for recognizing conspecific and \nheterospecific vocal emotions.