A study about the automatic recognition of the anxiety emotional state using Emo-DB

The main purpose of this paper is to determine how well can be differentiated the anxiety /fear emotion. In the analysis it is using EmoDB which contains a total number of seven emotions: happiness, fury, sadness, neutral tones, anxiety, boredom and disgust. We do not used the Romanian Database SRoL because the anxiety state is not recorded at this moment. The results are encouraging, the recognition error being 32-35% when using all seven emotions of Emo-DB or 24-30% when trying to identify the anxiety compared with only four basic emotions: happiness, fury, sadness, and neutral tones. This study shows that it can be extracted features from the voice which can help to identify the depression people that manifest deep sadness and anxiety states.