A diversity-penalizing ensemble training method for deep learning.

Xiaohui Zhang, Daniel Povey, Sanjeev Khudanpur

Browse the full Interspeech paper archive.