Lightly supervised training for risk-based discriminative language models.

Akio Kobayashi, Takahiro Oku, Yuya Fujita, Shoei Sato

Browse the full Interspeech paper archive.