Reconstruction of missing features based on a low-rank assumption for robust speaker identification.
Christos TzagkarakisAthanasios MouchtarisPublished in: IISA (2014)
Keyphrases
- low rank
- missing data
- rank minimization
- low rank matrix
- speaker identification
- feature extraction
- linear combination
- matrix factorization
- feature vectors
- convex optimization
- noisy environments
- low level
- singular value decomposition
- semi supervised
- matrix completion
- gaussian mixture model
- missing values
- high dimensional data
- kernel matrix
- feature space
- speech recognition
- norm minimization
- support vector
- face recognition