Abstract
Speech emotion recognition is an interesting research area with different applications. Hybridization of spectral features for speech emotion recognition is discussed in this paper. An emotional speech database consisting of nine emotions is created for this experiment. Extreme Learning Machines (ELM) is used as the pattern classifier. An overall recognition accuracy of 82.22% could be achieved from this experiment.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have