Basis superposition precision matrix modelling for large vocabulary continuous speech recognition.

Khe Chai Sim, Mark J. F. Gales

Browse the full ICASSP paper archive.