Joint sequence training of phone and grapheme acoustic model based on multi-task learning deep neural networks.

Dongpeng Chen, Brian Mak, Sunil Sivadas

Browse the full Interspeech paper archive.