Robust regression fusion of GMM-UBM and GMM-SVM normalized scores using G729 bit-stream for speaker recognition over IP.
Dalila YessadAbderrahmane AmrouchePublished in: Int. J. Speech Technol. (2014)
Keyphrases
- speaker recognition
- gaussian mixture model
- bitstream
- robust regression
- speaker identification
- feature vectors
- speaker verification
- density estimation
- mixture model
- vector quantization
- support vector
- coding scheme
- probabilistic neural network
- feature space
- speech signal
- bit rate
- speech recognition
- maximum likelihood
- feature selection
- knn
- machine learning
- support vector machine svm
- em algorithm
- linear regression
- multiresolution
- ls svm
- noisy environments
- support vector machine
- feature extraction