Facing Realism in Spontaneous Emotion Recognition from Speech: Feature Enhancement by Autoencoder with LSTM Neural Networks.
Zixing ZhangFabien RingevalJing HanJun DengErik MarchiBjörn W. SchullerPublished in: INTERSPEECH (2016)
Keyphrases
- emotion recognition
- neural network
- emotional speech
- audio visual
- recurrent neural networks
- human computer interaction
- facial expressions
- emotion classification
- speaker verification
- sentiment analysis
- pattern recognition
- emotional state
- information fusion
- artificial neural networks
- feature selection
- image features
- multi modal
- facial images
- back propagation
- multilayer perceptron
- feature set
- text mining
- feature vectors
- computer vision