Human Speech Emotion Recognition Using Artificial Neural Networks Technique

Speech Emotion Recognition is an essential research analysis for utilizing interconnection with embedded systems to introduce human interference technology. In this study, we proposed a model to recognize emotion from speech using the librosa and sklearn libraries from the dataset. The Mel Frequency Cepstral Coefficient (MFCC), MEL-SPECTROGRAM, Feature Extraction approach is used to identify different emotions in speech. Eight distinct emotional states, including neutral, calm, happy, sad, angry, fearful, disgusted, and surprised, are identified using the MLP classifier in “Native Languages”. The total findings of the studies conducted showed that the technique of utilizing the RAVDESS database in all classifiers obtained an accuracy of 53.89% in eight speaker-independent cases. The percentage of recognition comparison Datasets is higher than any other emotion. It is increasing efficiency with abate class of emotion.