Enhancing sparsity in linear prediction of speech by iteratively reweighted 1-norm minimization.
Daniele GiacobelloMads Græsbøll ChristensenManohar N. MurthiSøren Holdt JensenMarc MoonenPublished in: ICASSP (2010)
Keyphrases
- linear prediction
- norm minimization
- sparse representation
- speech signal
- convex optimization
- image coding
- prediction error
- sparse coding
- lossless compression
- vocal tract
- subband
- face recognition
- high dimensional
- autoregressive
- matrix completion
- negative matrix factorization
- unsupervised feature selection
- speech recognition
- interior point methods
- image representation
- dimensionality reduction
- filter bank
- image patches
- image restoration
- image classification
- image retrieval
- machine learning
- high dimensional data
- non stationary
- spectral analysis
- bit rate
- linear programming
- feature vectors
- information retrieval