Non-uniform unit parsing for SSS-LR continuous speech recognition.

Harald Singer, Jun-ichi Takami, Shoichi Matsunaga

Browse the full ICASSP paper archive.