Direct conversion from facial myoelectric signals to speech using Deep Neural Networks.
Lorenz DienerMatthias JankeTanja SchultzPublished in: IJCNN (2015)
Keyphrases
- neural network
- emotion recognition
- acoustic signals
- fundamental frequency
- signal processing
- audio signals
- pattern recognition
- artificial neural networks
- speech recognition
- facial expressions
- human faces
- landmark detection
- noisy environments
- facial animation
- facial motion
- facial images
- acoustic signal
- face recognition
- genetic algorithm
- cepstral features
- multilayer perceptron
- broadcast news
- fuzzy logic
- back propagation
- facial features
- recognition engine
- text to speech
- spectral analysis
- emotional state
- automatic speech recognition
- recurrent neural networks
- spoken language
- dialogue system
- feed forward
- neural network model
- activation function
- fault diagnosis
- audio visual
- language model
- face images
- facial expression recognition