Unsupervised acoustic and language model training with small amounts of labelled data.

Scott Novotney, Richard M. Schwartz, Jeff Z. Ma

Browse the full ICASSP paper archive.