Emotion recognition of speech based on RNN

Emotion Recognition has various methods. Mainly, it can be performed by visual or aural method. Any boy's expression face. informed others of his emotion. And, when people are talking over the telephone, they can know the opposite person's emotion by only sound data. From this point, we know that it is possible to recognize people's emotion by only sound data. In this paper, we use the pitch of speech as a main feature. And, as the most important thing, we define features of 4-emotions (normal, angry, laugh, surprise) in pitch analysis. And, based on this feature pattern, we implement a simulator by VC++. First of all, this simulator is composed of 'Generation of individuals', 'RNN', 'Evaluation'. And, using the result from learning part of this simulator, we can get results applied to other speech data (excepting for learning data). In detail, Each module uses the following method. First, 'generation of individuals'-part uses (1+100)-ES and (1+1)-ES (that is, random). Thus, we observe the comparison result of both methods. Of course, then, we select the best way. Second, 'RNN(Recurrent Neural Network)'-part is composed of 7-nodes. That is, 1-input node, 2-hidden layer nodes, 4-output nodes. Selection of this structure depends on the characteristics of sequentially inputted speech data. Third, 'evaluation'-part is very important This part is the cause of the extraction speed and satisfaction degree of result Then we implement a simulator from above modules. And, applied other speech data, we observe the result of recognition..