Structural MAP adaptation in GMM-supervector based speaker recognition.

Marc Ferras Koichi Shinoda Sadaoki Furui

Published in: ICASSP (2011)

Keyphrases

speaker recognition
gaussian mixture model
mixture model
speaker identification
feature vectors
em algorithm
maximum a posteriori
maximum likelihood
expectation maximization
vector quantization
multi modal
unsupervised learning
mel frequency cepstral coefficients