A high speed unsupervised speaker retrieval using vector quantization and second-order statistics
Konstantin BiatovPublished in: CoRR (2010)
Keyphrases
- vector quantization
- high speed
- speaker recognition
- image compression
- vector quantizer
- fractal image compression
- reduced complexity
- input vector
- fractal image coding
- neural gas
- distortion measure
- joint source and channel coding
- unsupervised learning
- multiscale
- vector quantized
- competitive learning
- audio visual
- speaker verification
- supervised learning
- image processing
- codebook generation
- neural network