Using a large vocabulary continuous speech recognizer for a constrained domain with limited training.

Man-Hung Siu, Michael Jonas, Herbert Gish

Browse the full ICASSP paper archive.