Audio/Speech Coding Based on the Perceptual Sparse Representation of the Signal with DAE Neural Network Quantizer and Near-End Listening Enhancement.
Vadzim HerasimovichAlexey A. PetrovskyVladislav AvramovAlexander A. PetrovskyPublished in: MISSI (2018)
Keyphrases
- sparse representation
- signal processing
- coding scheme
- neural network
- compressive sensing
- audio stream
- audio visual
- image processing
- dictionary learning
- pattern recognition
- video signals
- sparse coding
- broadcast news
- speech recognition
- vector quantization
- speaker identification
- random projections
- face recognition
- reconstruction error
- bitstream
- joint optimization
- audio features
- compressed sensing
- transform coefficients
- image patches
- test images
- audio signal
- speech signal
- image compression
- quantization scheme
- matching pursuit
- human visual system
- sparse signal representation
- high dimensional data
- image classification
- sparsity constraints
- image coding
- filter bank
- natural images
- training set
- video sequences
- image segmentation
- computer vision