MFCC-based Recurrent Neural Network for automatic clinical depression recognition and assessment from speech.
Emna RejaibiAli KomatyFabrice MériaudeauSaid AgrebiAlice OthmaniPublished in: Biomed. Signal Process. Control. (2022)
Keyphrases
- recurrent neural networks
- speech recognition
- speech signal
- recognition engine
- feature extraction
- speech recognition systems
- noisy speech
- neural network
- mel frequency cepstral coefficients
- speaker identification
- speech corpus
- complex valued
- automatic speech recognition systems
- feed forward
- cepstral coefficients
- recurrent networks
- speech music discrimination
- recognition rate
- pattern recognition
- speech enhancement
- echo state networks
- feedforward neural networks
- hidden markov models
- noisy environments
- reservoir computing
- speech synthesis
- speech sounds
- long short term memory
- handwriting recognition
- object recognition
- artificial neural networks
- recognition algorithm
- hidden layer
- real valued
- gaussian mixture model
- principal component analysis
- neural model
- neural network structure
- fuzzy neural network
- audio visual
- automatic speech recognition
- vocal tract
- feature vectors
- linear predictive
- spectral features
- decision making
- learning algorithm