Towards the use of full covariance models for missing data speaker recognition.

Marco Kühne Daniel Pullella Roberto Togneri Sven Nordholm

Published in: ICASSP (2008)

Keyphrases

missing data
speaker recognition
incomplete data
missing values
vector quantization
multiple imputation
training data
pairwise
gaussian mixture model
matrix factorization
neural network
pattern recognition
model selection
linear combination