MFCC-based Recurrent Neural Network for Automatic Clinical Depression Recognition and Assessment from Speech.
Emna RejaibiAli KomatyFabrice MériaudeauSaid AgrebiAlice OthmaniPublished in: CoRR (2019)
Keyphrases
- recurrent neural networks
- speech recognition
- speech signal
- recognition engine
- speech recognition systems
- feature extraction
- noisy speech
- cepstral coefficients
- neural network
- mel frequency cepstral coefficients
- feed forward
- automatic speech recognition
- complex valued
- recurrent networks
- pattern recognition
- speaker identification
- speech sounds
- recognition rate
- object recognition
- automatic speech recognition systems
- artificial neural networks
- speech corpus
- speaker recognition
- recognition algorithm
- handwriting recognition
- reservoir computing
- neural model
- hidden layer
- hidden markov models
- linear predictive
- long short term memory
- noisy environments
- speaker diarization
- nonlinear dynamic systems
- speech synthesis
- acoustic features
- speaker verification
- feedforward neural networks
- support vector machine
- support vector
- automatic transcription
- speech music discrimination
- high risk