Optimal selection of bitstream features for compressed-domain automatic speaker recognition.
Matteo PetraccaAntonio ServettiJuan Carlos De MartinPublished in: EUSIPCO (2006)
Keyphrases
- compressed domain
- bitstream
- optimal selection
- speaker recognition
- bit rate
- coding scheme
- compressed video
- compression algorithm
- video analysis
- video quality
- compressed images
- motion vectors
- dct domain
- feature extraction
- dct coefficients
- gaussian mixture model
- pixel domain
- feature vectors
- wavelet coefficients
- motion compensation
- speaker identification
- speaker verification
- inter frame
- neural network
- video data
- classification accuracy
- feature space
- video sequences
- multiscale
- image segmentation
- machine learning