Discriminative Feature Extraction Based on Sequential Variational Autoencoder for Speaker Recognition.
Takenori YoshimuraNatsumi KoikeKei HashimotoKeiichiro OuraYoshihiko NankakuKeiichi TokudaPublished in: APSIPA (2018)
Keyphrases
- speaker recognition
- feature extraction
- speaker identification
- gaussian mixture model
- vector quantization
- probabilistic neural network
- speaker verification
- mel frequency cepstral coefficients
- feature vectors
- feature space
- pattern classification
- image classification
- wavelet transform
- feature selection
- image processing
- image segmentation
- face recognition
- discriminant analysis
- linear discriminant analysis
- pattern recognition
- extracting features
- speech signal
- extracted features
- speech recognition
- support vector machine svm
- dimensionality reduction
- principal component analysis
- texture features
- dimension reduction
- em algorithm
- language identification
- image compression
- palmprint
- mixture model
- unsupervised learning
- semi supervised