Using smoothed K-TSS language models in continuous speech recognition.

Amparo Varona, Ins Torres

Browse the full ICASSP paper archive.