Robust Audio-visual Speech Recognition Using Bimodal Dfsmn with Multi-condition Training and Dropout Regularization.

Shiliang Zhang, Ming Lei, Bin Ma, Lei Xie

Browse the full ICASSP paper archive.