Optimal feature selection based speech emotion recognition using two-stream deep convolutional neural network.
MustaqeemSoonil KwonPublished in: Int. J. Intell. Syst. (2021)
Keyphrases
- feature selection
- convolutional neural network
- feature subset
- optimal solution
- data streams
- emotion recognition
- dynamic programming
- speech recognition
- machine learning
- multi stream
- text classification
- feature set
- text to speech synthesis
- audio stream
- high dimensionality
- text categorization
- worst case
- facial expressions
- support vector
- feature extraction
- neural network
- real time
- sliding window
- speech signal
- emotional state
- object recognition
- genetic algorithm